Louisiana State University Louisiana State University

LSU Digital Commons LSU Digital Commons

Faculty Publications Department of Geology and Geophysics

1-1-2020

Large-scale mass wasting on the miocene continental margin of Large-scale mass wasting on the miocene continental margin of

Western India Western India

Sarah K. Dailey Louisiana State University

Peter D. Clift Louisiana State University

Denise K. Kulhanek Texas A&M University

Jerzy Blusztajn Woods Hole Oceanographic Institution

Claire M. Routledge University College London

See next page for additional authors

Follow this and additional works at: https://digitalcommons.lsu.edu/geo_pubs

Recommended Citation Recommended Citation Dailey, S., Clift, P., Kulhanek, D., Blusztajn, J., Routledge, C., Calvès, G., O'Sullivan, P., Jonell, T., Pandey, D., Andò, S., Coletti, G., Zhou, P., Li, Y., Neubeck, N., Bendle, J., Aharonovich, S., Griffith, E., Gurumurthy, G., Hahn, A., Iwai, M., Khim, B., Kumar, A., Kumar, A., Liddy, H., Lu, H., Lyle, M., Mishra, R., Radhakrishna, T., Saraswat, R., Saxena, R., Scardia, G., Sharma, G., & Singh, A. (2020). Large-scale mass wasting on the miocene continental margin of Western India. Bulletin of the Geological Society of America, 132 (1-2), 85-112. https://doi.org/10.1130/B35158.1

This Article is brought to you for free and open access by the Department of Geology and Geophysics at LSU Digital Commons. It has been accepted for inclusion in Faculty Publications by an authorized administrator of LSU Digital Commons. For more information, please contact ir@lsu.edu.

Authors Authors Sarah K. Dailey, Peter D. Clift, Denise K. Kulhanek, Jerzy Blusztajn, Claire M. Routledge, GérÔme Calvès, Paul O'Sullivan, Tara N. Jonell, Dhananjai K. Pandey, Sergio Andò, Giovanni Coletti, Peng Zhou, Yuting Li, Nikki E. Neubeck, James A.P. Bendle, Sophia Aharonovich, Elizabeth M. Griffith, Gundiga P. Gurumurthy, Annette Hahn, Masao Iwai, Boo Keun Khim, Anil Kumar, A. Ganesh Kumar, Hannah M. Liddy, Huayu Lu, Mitchell W. Lyle, Ravi Mishra, Tallavajhala Radhakrishna, Rajeev Saraswat, Rakesh Saxena, Giancarlo Scardia, Girish K. Sharma, and Arun D. Singh

This article is available at LSU Digital Commons: https://digitalcommons.lsu.edu/geo_pubs/293

UC San DiegoUC San Diego Previously Published Works

TitleLarge-scale mass wasting on the Miocene continental margin of western India

Permalinkhttps://escholarship.org/uc/item/7174d6xp

JournalGEOLOGICAL SOCIETY OF AMERICA BULLETIN, 132(1-2)

ISSN0016-7606

AuthorsDailey, Sarah KClift, Peter DKulhanek, Denise Ket al.

Publication Date2020

DOI10.1130/B35158.1 Peer reviewed

eScholarship.org Powered by the California Digital LibraryUniversity of California

1

Large-scale Mass Wasting on the Miocene Continental Margin of Western 1

India 2

3

Sarah K. Dailey1, Peter D. Clift1, Denise K. Kulhanek2, Jerzy Blusztajn3, Claire M. Routledge4, 4

Gérôme Calvès5, Paul O’Sullivan6, Tara N. Jonell7, Dhananjai K. Pandey8, Sergio Andò9, 5

Giovanni Coletti9, Peng Zhou1, Yuting Li1, Nikki E. Neubeck1, James A.P. Bendle10, Sophia 6

Bratenkov11, Elizabeth M. Griffith12, Gundiga P. Gurumurthy13, Annette Hahn14, Masao Iwai15, 7

Boo-Keun Khim16, Anil Kumar17, A. Ganesh Kumar18, Hannah M. Liddy19, Huayu Lu20, 8

Mitchell W. Lyle21, Ravi Mishra8, Tallavajhala Radhakrishna22, Rajeev Saraswat23, Rakesh 9

Saxena24, Giancarlo Scardia25, Girish K. Sharma26, Arun D. Singh27, Stephan Steinke28, Kenta 10

Suzuki29, Lisa Tauxe30, Manish Tiwari9, Zhaokai Xu31, and Zhaojie Yu32 11

12

1 - Department of Geology and Geophysics, Louisiana State University, E253 Howe-Russell-Kniffen 13 Geoscience Complex, Baton Rouge LA 70803, USA 14 2 - International Ocean Discovery Program, Texas A&M University, 1000 Discovery Drive, College 15 Station, TX 77845, USA 16 3 – Department of Geology and Geophysics, Woods Hole Oceanographic Institution, Woods Hole, MA 17 02543, USA 18 4 - Department of Earth Sciences, University College London, Gower Street, London, WC1E 6BT, 19 United Kingdom 20 5 - Université Toulouse 3, Paul Sabatier, Géosciences Environnement Toulouse, 14 avenue Edouard 21 Belin, 31400, Toulouse, France 22 6 - GeoSep Services, 1521 Pine Cone Road, Moscow, Idaho 83843 USA 23 7 - School of Earth and Environmental Sciences, University of Queensland, QLD 4072, Australia 24 8 - National Centre for Antarctic and Ocean Research (NCAOR), Vasco da Gama, Goa 403804, India 25 9 - Department of Earth and Environmental Sciences, University of Milano Bicocca, Piazza della Scienza 26 4, 20126 Milan, Italy 27 10 - School of Geography, Earth and Environmental Sciences, University of Birmingham, Edgbaston, 28 Birmingham B15 2TT, United Kingdom 29 11 - Department of Earth and Planetary Sciences, Macquarie University, 202/1 Botany Rd., Sydney 30 NSW 2017, Australia 31

2

12 - School of Earth Sciences, Ohio State University, 275 Mendenhall Lab, 25 South Oval Mall, 32 Columbus OH, 43210, USA 33 13 - Manipal Centre for Natural Sciences, Manipal University, Manipal 576104, India 34 14 – MARUM, University of Bremen, Leobener Strasse, Bremen 28359, Germany 35 15 - Department of Natural Environmental Science, Kochi University, 2-5-1 Akebono-cho, Kochi 780-36 8520, Japan 37 16 - Division of Earth Environmental System, Pusan National University, Jangjeon-dong, Geumjeong-38 gu, Busan 609-73, Korea 39 17 - Department of Science and Technology Wadia Institute of Himalayan Geology, 33 GMS Road, 40 Dehradun, Uttrakhand 248001, India 41 18 - Marine Biotechnology Department, National Institute of Ocean Technology, Velachery-Tambaram 42 Main Road, Pallikaranai, Chennai 600100, India 43 19 - The Earth Institute, Columbia University, Hogan Hall, 2910 Broadway, Level A, New York NY 44 10025, USA 45 20 - School of Geographical and Oceanographical Sciences, Nanjing University, 163 Xianlin Avenue, 46 Nanjing 210023, P.R. China 47 21 - College of Earth, Ocean and Atmospheric Sciences, Oregon State University, 104 CEOAS 48 Administration Building, Corvallis OR 97331, USA 49 22 - Geosciences Division, National Centre for Earth Science Studies, Aakkulam Trivandrum 695031, 50 India 51 23 - Geological Oceanography Division, National Institute of Oceanography, Dona Paula, Goa 403004, 52 India 53 24 – ONGC, 11 High, Bandra-Sion Link Road, Mumbai 400017, India 54 25 - Instituto de Geociências e Ciências Exatas, Universidade Estadual Paulista, 1515 Avenida 24-A, Rio 55 Claro SP 13506-900, Brazil 56 26 - Department of Geology, Kumaun University, Nainital 263002, India 57 27 - Department of Geology, Banaras Hindu University, Varanasi Uttar Pradesh 221005, India 58 28 - Department of Geological Oceanography and State Key Laboratory of Marine Environmental 59 Science, Xiamen University, Xiamen 361102, P.R. China 60 29 - Graduate School of Environmental Science, Hokkaido University, N10W5, Kita-ku, Sapporo 060-61 0810, Japan 62 30 - Scripps Institution of Oceanography, 9500 Gilman Drive, La Jolla CA 92093-0220, USA 63 31 - Key Laboratory of Marine Geology and Environment, Institute of Oceanology, Chinese Academy of 64 Sciences, 7 Nanhai Road, Qingdao Shandong 266071, P.R. China 65 32 - University of Paris XI (Orsay), Bâtiment 504, Orsay Cedex 91405, France 66 67 68

3

Abstract 69

A giant mass transport complex was recently discovered in the eastern Arabian Sea, 70

exceeding in volume all but one other known complex on passive margins worldwide. The 71

complex, named the Nataraja Slide, was drilled by International Ocean Discovery Program 72

(IODP) Expedition 355 in two locations where it is ~300 m (Site U1456) and ~200 m thick (Site 73

U1457). The top of this mass transport complex is defined by the presence of both reworked 74

microfossil assemblages and deformation structures, such as folding and faulting. The deposit 75

consists of two main phases of mass wasting, each which consists of smaller pulses, with 76

generally fining-upward cycles, all emplaced just prior to 10.8 Ma. The base of the deposit at 77

each site is composed largely of matrix-supported carbonate breccia that is interpreted as the 78

product of debris flows. In the first phase, these breccias alternate with well-sorted calcarenites 79

deposited from a high energy current, coherent limestone blocks that are derived directly from 80

the Indian continental margin, and a few clastic mudstone beds. In the second phase, at the top of 81

the deposit, muddy turbidites dominate and become increasingly more siliciclastic. At Site 82

U1456, where both phases are seen, a 20 m section of hemipelagic mudstone is present, overlain 83

by a ~40 m thick section of calcarenite and slumped interbedded mud and siltstone. Bulk 84

sediment geochemistry, heavy- mineral analysis, clay mineralogy, isotope geochemistry, and 85

detrital zircon U-Pb ages constrain the provenance of the clastic, muddy material to being 86

reworked Indus-derived sediment, with input from western Indian rivers (e.g., Narmada and 87

Tapti Rivers), and some material from the Deccan Traps. The carbonate blocks found within the 88

breccias are shallow-water limestones from the outer western Indian continental shelf that was 89

oversteepened from enhanced clastic sediment delivery during the mid-Miocene. The final 90

emplacement of the material was likely related to seismicity as there are modern analogues for 91

4

intraplate earthquakes close to the source of the slide. Although we hypothesize this area is at 92

low risk for future mass wasting events, it should be noted that other oversteepened continental 93

margins around the world could be at risk for mass failure as large as the Nataraja Slide. 94

95

INTRODUCTION 96

Large-scale mass wasting of continental margins is an important process in controlling 97

the geomorphology of continental slopes fringing all ocean basins (Coleman and Prior, 1988). 98

The scale of large mass transport complexes (MTCs) makes them significant as geohazards, 99

directly through mass wasting (Dan et al., 2007; Yamada et al., 2012), by generating tsunamis 100

(Tappin et al., 2001), as well as posing risks for seafloor infrastructure such as oil and gas 101

platforms, pipelines (Bea et al., 1983), and communication cables (Hsu et al., 2008). Moreover, 102

the emplacement of MTCs can have significant influence on the stratigraphy of deep ocean 103

basins, as well as for the continental margin from which it was derived. 104

Although the largest mass transport deposits are associated with active margins (Burg et 105

al., 2008), where earthquakes are more common and can act as triggers for emplacement, passive 106

margins are also recognized to host some of the largest gravitational collapses in the modern 107

oceans (Embley and Jacobi, 1977). Seismic surveying in the eastern Arabian Sea offshore 108

western India has identified one of the largest such complexes, totaling around 19,000 km³ 109

(Calvès et al., 2015). Mapping of the deposit by seismic methods suggests that it may be up to 110

800 m thick in places (Calvès et al., 2015). In 2015 this deposit was drilled by International 111

Ocean Discovery Program (IODP) during Expedition 355. During the expedition, the MTC was 112

sampled on its southern edge, where the thicknesses were considerably thinner (Pandey et al., 113

2016c)(Fig. 1). The deposit, named the Nataraja Slide, shows substantial run out from its inferred 114

5

source regions offshore Saurashtra (Fig. 1), being emplaced ~500 km into the Indian Ocean. In 115

this study, we examine the sedimentary rocks recovered by IODP in order to infer the 116

depositional mechanisms active during emplacement. We further make inferences about what 117

processes triggered its formation, which is dated as being just before 10.8 Ma (Pandey et al., 118

2016a). Are MTCs of this magnitude formed by the same processes that we see at much smaller 119

scales, or are these mega-scale complexes unique in their modes of emplacement and triggers? 120

Given the profound potential geohazards for human settlements in coastal regions, understanding 121

the origins and impacts of the Nataraja Slide MTC are of both great scientific and societal 122

significance. 123

124

GEOLOGY OF LARGE MTCS 125

Mass transport complexes are an extreme form of gravity induced sediment transport 126

(Hampton et al., 1996). Most submarine gravity driven sediment transport involves redeposition 127

of individual sediment particles suspended in water (e.g., in a turbidity current) or as a fluidized 128

sediment suspension (e.g., a debris flow or mud flow)(Pickering et al., 1986; Talling et al., 129

2012). Sediment may also be mobilized when the proportion of water is very low, such as a 130

slow- moving sediment grain flow or creep (Carter, 1975; Lowe, 1976). However, large volumes 131

of material can also be transported rapidly (hours to days) in the form of slope failures where 132

coherent masses of material can be transported by sliding, rolling, falling, and/or slumping 133

(Coleman and Prior, 1988). Slumps involve displacement of a stratigraphic package above a 134

concave-upward detachment surface and can leave the slumped material in a relatively 135

undisturbed state after removal from an area that then shows an arcuate scar (Hampton et al., 136

1996; Moore, 1961). Slumps differ from slides in that motion is along a pre-existing weakness, 137

6

such as a bedding plane or joint surface, but the displaced package can move as a coherent mass, 138

or can be become disaggregated depending on the length and speed of transport. Significant 139

progress has been made in understanding mass transport through outcrop studies, such as the 140

Carboniferous (Pennsylvanian) Ross Slide of Ireland (Martinsen and Bakken, 1990; Strachan, 141

2002), the Eocene of the Pyrenean foreland basin (Farrell, 1984), and the Pliocene of Sicily 142

(Trincardi and Argnani, 1990). In all examples, each MTC was emplaced over a sharply defined 143

basal décollement once the deposit reached the lower slope after erosive mass wasting of the 144

steeper upper slope. 145

The geometry and internal structure of any gravitationally driven slump, slide or debris 146

flow reflect the mechanism of failure and the morphology of the slope where the transport occurs 147

(Lucente and Pini, 2003). The style of deformation and the mode of transport are controlled by 148

sediment and rock rheology that in turn are dependent on the lithology and strain rate. For this 149

reason, the largest MTCs are different from shallow debris flows and slumps because they 150

incorporate both lithified and unconsolidated materials. There are few exposures of very large 151

MTCs and those in the oceans are hard to access, especially through drilling. MTCs are often 152

seismically homogeneous (Vardy et al., 2010) but can show important changes in sediment 153

facies with depth and with distance from their source. For example, swath bathymetric mapping 154

of the Ebro margin in the western Mediterranean featuring the pre-11 ka BIG’95 Slide shows 155

that only finer sediments have reached the most distal areas, yet coherent rafts of continental 156

margin sedimentary rock are seen at the base of the slope (Lastras et al., 2004). Analysis of the 157

geometry and distribution of sedimentary facies and structures can be used to reconstruct the 158

evolving sedimentary and deformational strain history of any individual MTC. By doing so, it is 159

possible to derive a kinematic model of emplacement that can be compared with other examples. 160

7

The Storegga Slide in offshore Norway is one of the best studied large-volume mass 161

transport complex. This MTC is entirely siliciclastic and its generation has been linked to sliding 162

on contourite sand and silts that became overpressured as a result of rapid burial by glacial 163

maximum aged debris flow sediments (Bryn et al., 2005). However, rapid sedimentation on any 164

clastic margin receiving sediment from the continent would provide weak layers on which 165

sliding could occur. Overpressuring has also been linked to growth and migration of silica 166

diagenetic fronts (Davies and Clark, 2006). Slope oversteepening increases the chances of mass 167

wasting simply by the consequence of rapid sediment delivery, although the tendency may be 168

heightened by the pre-existing basement structure of the continental margin (Lastras et al., 2004). 169

Slope oversteepening by itself cannot explain large-scale mass wasting because giant MTCs on 170

European continental margins are mostly associated with low gradient glacial margins. In 171

contrast, turbidity currents appear to dominate on steeper non-glacial margins which might 172

otherwise be expected to suffer mass wasting due to their gradient (Leynaud et al., 2009). In 173

these cases, differences in the sediment types and the timing of sediment delivery favor 174

gravitational instabilities at different times, with non-glaciated margins tending to mass waste 175

more during sealevel lowstands, where the opposite more often occurs on glaciated margins. 176

Modelling indicates that continental margins with more cohesive clay-rich sediments tend to 177

experience coherent sliding more frequently than sand-rich margins whose gravitational slides 178

tend to disintegrate into turbidity currents (Elverhoi et al., 2010). 179

The triggering of MTC emplacement can be attributed to a number of potential processes, 180

including seismicity (Moernaut et al., 2007; Piper et al., 1985), volcanic eruptions (Carracedo, 181

1999) and meteorite impacts (Klaus et al., 2000; Parnell, 2008). Dissociation of gas hydrates 182

during times of warming seawater could have aided liquefaction in the case of Storegga Slide 183

8

(Mienert et al., 2005), with seismicity possibly related to post-glacial isostatic rebound providing 184

the final impetus for redeposition (Evans et al., 2002). In the eastern Mediterranean Sea, MTC 185

emplacement has also be linked to biogenic gas and slope oversteepening acting individually or 186

in tandem with one another (Frey Martinez et al., 2005). 187

Mechanisms for MTC emplacement differ between clastic and carbonate margins. This is 188

because carbonate sediment production occurs in situ and can result in steep platform margins, 189

sometimes almost vertically where reef complexes develop in outer shelf areas. Carbonate 190

production is strongly linked to sealevel and was fastest when sealevel was high after the onset 191

of Northern Hemispheric Glaciation (NHG, ~2.4 Ma)(Schlager et al., 1994). Many carbonate 192

MTCs are linked to platform margin collapse and result in deposits with numerous coherent 193

blocks suspended within a more fluidized matrix. Seismic mapping around the Great Bahama 194

Bank has identified coherent Plio-Pleistocene sedimentary rock rafts 0.5–2.0 km in length, 0.3–195

1.5 km in width, and 50 m in thickness (Principaud et al., 2015). Adjacent deposits have also 196

been observed on the Florida margin (Mullins et al., 1986), as well as offshore Nicaragua (Hine, 197

1992), all with a similar Plio-Pleistocene age. Plio-Pleistocene MTCs are larger than most known 198

older examples because the rapidly changing sealevel since the start of the NHG enhanced 199

carbonate production and induced gravitational instability as sealevel rose and fell (Schlager et 200

al., 1994). Among these older deposits, only the Cretaceous Ayabacas MTC of Peru is 201

noteworthy for its large volume, long run out and presence of slide blocks measuring kilometers 202

in length (Callot et al., 2008). 203

204

GEOLOGICAL SETTING 205

9

The Nataraja Slide lies within the Laxmi Basin offshore the western continental margin 206

of India (Fig. 1A and B). The Laxmi Basin is separated from the main Arabian Basin by the 207

Laxmi Ridge (Fig. 1). The Laxmi Basin is a rift basin that formed between India and the Laxmi 208

Ridge prior to the opening of the main Arabian Sea in the early Paleocene (Bhattacharya et al., 209

1994), where the ridge is generally interpreted to be a rifted fragment of Indian continental crust 210

(Pandey et al., 1995). The age of rifting is somewhat controversial, but likely just predates the 211

emplacement of the Deccan Traps flood basalts in the latest Cretaceous, based on analysis of 212

magnetic anomalies (Bhattacharya et al., 1994) and the geochemistry of the basalts sampled at 213

IODP Site U1457 (Pandey et al., 2016b). The sediments in the Laxmi Basin can be divided into 214

three major units described below. The oldest, dated as Lower Paleocene, largely comprises red-215

brown mudstones eroded from peninsular India and sampled at IODP Site U1457 (Pandey et al., 216

2016b). These deposits are overlain by the Nataraja Slide and by younger distal turbidite 217

sandstones and siltstones, as well as hemipelagic mudstones that form the Indus submarine fan. 218

These latter sediments were supplied through the Indus River via erosion from the western 219

Himalaya and Karakoram (Pandey et al., 2016c). The age of the Indus Fan in the Laxmi Basin is 220

not well defined, although within the main Arabian basin the fan is typically considered to date 221

from at least 45 Ma, continuing to the present time (Clift et al., 2001). It is within these deposits 222

that the Nataraja Slide (MTC) was emplaced just before 10.8 Ma. 223

Towards the east, the Laxmi Basin is bounded by the rifted passive margin of India, 224

which has been supplied by sediment from the erosion of the peninsula via a number of 225

significant rivers that drain towards the west (e.g., Mahi, Tapti, and Narmada). Oil exploration 226

drilling has furthermore identified significant repeated buildups of carbonate on the shelf, 227

especially towards the shelf edge where the supply of clastic material was more limited (Rao and 228

10

Talukdar, 1980; Wandrey, 2004). It is generally presumed that extensional deformation in the 229

area ceased after the rifting that formed the Laxmi Basin. The area has been largely seismically 230

inactive except towards the north where the Rann of Kutch forms an active structure within the 231

Indian Craton. This structure is linked to flexure of the plate as a result of the collision between 232

India and Asia (Bilham et al., 2003; Biswas, 2005), presumed to have started in the Eocene 233

(Najman et al., 2010) or even earlier (DeCelles et al., 2014). Towards the north, the Indian 234

peninsula is cut by the NE-SW trending Cambay Basin which formed as an initial early 235

Cretaceous rift that was then reactivated in the Cenozoic and experienced significant inversion in 236

the early Miocene (Chowdhary, 2004). 237

The MTC run-out distance is estimated to be about 550 km, with a length of 338 km and 238

a maximum width of 193 km (Calvès et al., 2015). Prior work on the Nataraja Slide found this 239

MTC to be acoustically homogenous in seismic lines, with few identified rafts present, and to 240

have a flat, rather than significantly angular erosive base over older deposits (Fig. 2)(Calvès et 241

al., 2015; Pandey et al., 2016c). However, closer inspection in the vicinity of the drilling sites 242

finds this is not always the case. In the case of IODP Site U1456 where the slide is somewhat 243

thicker, there is a significant missing section of submarine fan turbidites from ~15.6 to 10.8 Ma 244

(Pandey et al., 2016a). In that area the upper part of the deposit appears to be more acoustically 245

washed out and homogenous, but the lower regions are marked by strong reflections that show 246

limited lateral continuity suggestive of some internal structure within the deposit. This raises the 247

possibility that this is not simply a single depositional package (Fig. 2). Such strong reflections 248

are reminiscent of coherent slide blocks seen in seismic images of other MTCs (Gamboa et al., 249

2012; Krastel et al., 2012; Principaud et al., 2015). The same is not true at the more distal Site 250

11

U1457 location where the MTC onlaps the Laxmi Ridge and its acoustic character is more 251

uniform. 252

253

METHODS 254

Sedimentary cores were collected and initially described during IODP Expedition 355, 255

but several cores are re-examined in order to obtain more detailed descriptions of critical 256

sedimentary structures and facies. In addition to preparing sedimentary logs designed to 257

highlight the contrasting sedimentary facies, samples for sediment petrography were examined to 258

allow investigation into the different sediment types at both the macro and microscopic scale. 259

These methods allowed us to better define the depositional processes that operated during 260

Nataraja Slide emplacement and to provide constraints on the origin(s) of the MTC. 261

Geochemical methods were employed in order to further constrain the provenance of the 262

materials, and in particular, to verify the proposed western Indian continental margin source for 263

much of the MTC argued by Calvès et al. (2015). This approach is predicated on the fact that 264

source rocks of MTC deposits have different bulk geochemical compositions and that Himalayan 265

sources can be effectively discriminated from peninsular sources when considering provenance 266

due to different bedrock source compositions and contrasting chemical weathering histories. 267

Forty-four samples were selected for determination of major element composition, 268

together with select trace elements (Ni, Ba, V, Zr, Sc, Y, Sr). These were determined by 269

inductively coupled plasma emission spectrometry (ICP-ES) at Boston University, with precision 270

quantified to be better than 2% of the measured value for all elements. Accuracy was constrained 271

by analysis of certified Standard Reference Materials (BHVO-2) and results were accurate within 272

precision. Table 1 provides analyses of samples as well as repeated analyses of the standard. 273

12

The neodymium (Nd) isotope compositions of sediments are generally considered to be 274

minimally affected by chemical weathering, such that source terranes faithfully translate their 275

isotopic signature to eroded sediments (i.e., Goldstein et al. (1984)) and can be utilized for 276

sedimentary provenance studies. Strontium (Sr) isotopes are additionally considered, while 277

recognizing that Sr isotope compositions may be affected by chemical alteration largely during 278

transport across flood plains (Derry and France-Lanord, 1996). Together these isotopic systems 279

have a record of being powerful provenance proxies in the Arabian Sea (e.g., (Clift and 280

Blusztajn, 2005; Clift et al., 2008a)). Care was taken to decarbonate samples prior to analysis 281

with 20% acetic acid because Sr isotope compositions are strongly controlled by carbonate 282

compositions and this study targets the siliciclastic sediment compositions only. Decarbonation 283

lasted for six days until no further fizzing was observed when samples were exposed to 284

unreacted acid. Samples were washed by deionized water before being ground into powders. 285

Twenty-five samples were selected throughout the Nataraja Slide/MTC at Sites U1456 and 286

U1457 for the determination of 143Nd/144Nd and 87Sr/86Sr values. Isotopic compositions were 287

determined by Finnigan Neptune multi-collector inductively coupled plasma mass spectrometer 288

(MC-ICP-MS) at the Woods Hole Oceanographic Institute for both Nd and Sr isotopes. Nd and 289

Sr isotope analyses were corrected against La Jolla Nd standard 143Nd/144Nd=0.511847 and 290

NBS987 standard 87Sr/86Sr=0.710240. Procedural blanks were 20–25 pg for Sr and 50–70 pg for 291

Nd. We calculate the parameter ϵNd after (DePaolo and Wasserburg, 1976) using a 143Nd/144Nd 292

value of 0.512638 for the Chondritic Uniform Reservoir (CHUR) (Hamilton et al., 1983). 293

Results are presented in Table 2. 294

Heavy-mineral analysis was applied to study the mineralogy of the MTC deposits in 295

order to further constrain the source of the materials and to estimate the potential impact of 296

13

diagenetic dissolution. Sediment left after thin section preparation was gently crushed in water 297

with mortar and pestle and wet-sieved using a standard 500 m steel sieve and a special 298

handmade 15 m tissue-net sieve. A wide size window (15–500 m) was chosen to include a 299

large range of the size distribution (Garzanti et al., 2009). A gravimetric separation of dense 300

grains was achieved with a centrifuge using Na-polytungstate (density 2.90 g/cm3), and heavy 301

minerals recovered by partial freezing in liquid nitrogen. An appropriate amount of the dense 302

fraction thus obtained was split with a micro-riffle box and mounted with Canada balsam. Heavy 303

minerals were counted under a polarizing microscope with the area method (Mange and Maurer, 304

1992). Grains of uncertain character were systematically checked and identified by an inViaTM 305

Renishaw Raman spectrometer, equipped with a 532 nm laser and a 50x LWD objective (Andò 306

and Garzanti, 2014). Heavy-mineral and transparent-heavy-mineral concentrations (HMC and 307

tHMC indices of Garzanti and Andò (2007), representing fundamental parameters for 308

unravelling provenance and detecting hydraulic-sorting effects and diagenesis, allow us to 309

distinguish poor (tHMC < 1), and very rich (tHMC > 10) transparent-heavy-mineral suites. The 310

resulting assemblages were compared with those of modern sediments of the Tapti River 311

(sampled at 21°08’40.7” N, 72°44’08.1”E) and Indus River. Results are presented in Table 3. 312

U-Pb dating of detrital zircon has been widely used for provenance analysis in siliciclastic 313

systems because zircon is a common mineral in continental rocks of many compositions and is 314

chemically and mechanically resistant to weathering during transport (Carter and Bristow, 2003). 315

Furthermore, zircon has a closure temperature of ~750°C for the U/Pb isotope system (Hodges, 316

2003), making it very robust and unsusceptible to change during multiple stages of recycling. 317

Mineral separation and grain mounting were performed at GeoSep Services (GSS) Laboratory, 318

Moscow, ID. Only one sample was analyzed for zircon U-Pb dating because much of the core 319

14

lacked suitable layers for this method. Zircons were separated via hand picking and used for age 320

dating as described by Donelick et al. (2005). This process enhances the recovery of all possible 321

grain sizes while minimizing the potential loss of smaller grains within a sample by the use of 322

water-table devices. The method used by Donelick et al. (2005) further ensures the preservation 323

of complete grains by minimizing grain breakage and/or fracturing that can be associated with 324

traditional procedures of isolating individual grains from whole rock samples. Recovered zircon 325

were mostly medium silt to fine sand-sized grains. Epoxy wafers containing zircon grains for 326

laser ablation inductively coupled plasma mass spectrometry (LA-ICP-MS) were polished 327

manually using 3.0 μm and 0.3 μm Al2O3 slurries to expose internal zircon grain surfaces. The 328

polished grain surfaces were washed in 5.5 M HNO3 for 20 sec. at 21°C in order to clean the 329

surfaces prior to introduction into the laser system sample cell. 330

A total of 51 individual zircon grains were targeted for data collection using a New Wave 331

YP213 213 nm solid state laser ablation system with a 20 μm diameter laser spot size, 5 Hz laser 332

firing rate, and ultra-high purity He as the carrier gas. Isotopic analyses of the ablated zircons 333

were performed using a ThermoScientific Element 2 magnetic sector mass spectrometer using 334

high purity Ar as the plasma gas. Ages from the ratios 207Pb/235U, 206Pb/238U, and 207Pb/206Pb 335

were calculated for each data scan and checked for concordance. Concordance was defined as 336

overlap of all three ages at the 1σ level. If the number of concordant data scans for a spot was 337

greater than zero, the more precise age from the concordant-scan-weighted ratio 207Pb/235U, 338

206Pb/238U, or 207Pb/206Pb was chosen as the preferred age, and whichever exhibited the lower 339

relative error. If zero concordant data scans were observed, the common Pb-corrected age based 340

on isotopic sums of all acceptable scans was chosen as the preferred age. Results of zircon U-Pb 341

dating are shown in Table 4. 342

15

Clay mineralogy was examined for provenance purposes based on the concept that different 343

environmental conditions and source terranes can produce characteristic assemblages. This 344

allows us to separate material derived from the Indus River from material more closely linked to 345

peninsular India. Although there may have been some change in mineralogy during initial 346

diagenesis, the relatively shallow burial depths of these cores means that there is no significant 347

thermal diagenesis and we can consider the observed mineralogy to be largely representative of 348

that at the time of sedimentation. 349

Clay mineralogy was determined by using X-Ray Powder Diffraction (XRD) at Louisiana 350

State University using a Panalytical Empyrean X-Ray Diffractometer. Forty selected samples 351

within the MTC were soaked in water until there was no flocculation, with Na3PO4 added to de-352

flocculate when necessary. Samples were centrifuged for separation of the <2 μm material. Four 353

XRD patterns were generated from each oriented sample smear. The first pattern was collected 354

from the sample in air-dried conditions. The second XRD pattern was generated from a 355

glycolated sample after the slide was then placed in a desiccator with ethylene glycol for a 356

minimum of 8 h at 25°C. t. The third and fourth XRD datasets were collected after the sample 357

was subjected to heat treatments of 300°C for 1 h, and then 550°C for 1 h, respectively. XRD 358

analysis began immediately after glycolation, and immediately after the first heat treatment. In 359

this study we use the semi-quantitative method of Biscaye (1965) to estimate the clay 360

assemblage, which is based on peak-intensity factors determined from calculated XRD patterns 361

as measured by MACDIFF software. For clay minerals present in amounts >10 wt% uncertainty 362

is estimated as better than ±5 wt% at the 95% confidence level. Uncertainty of peak area 363

measurement based on repeated measurements is typically <5%. Data are presented as relative 364

concentrations of the total clay assemblage in Table 5. 365

16

366

DEFINING THE TOP AND BASE 367

Microfossil assemblages within the sediments provide constraints on the age of 368

emplacement. The oldest sediment overlying the MTC was dated at around 10.8 Ma based on 369

nannofossil assemblages and paleomagnetic stratigraphy (Pandey et al., 2016c). In Hole U1456D 370

the first appearance of Discoaster hamatus (10.55 Ma) marks the top of Zone NN8 (Pandey et 371

al., 2016a), while in Hole U1457C the interval 859.49–995.93 mbsf contains Catinaster coalitus, 372

which has a total age range of 9.69–10.89 Ma (Pandey et al., 2016b). The presence of Discoaster 373

bellus (first appearance at 10.40 Ma) within this interval also constrains the age to between 9.69 374

and 10.40 Ma. Much of the interval from 1009.21 to 1054.34 mbsf at Site U1457 contains a 375

mixture of different nannofossil species. 376

Above the MTC there is a coherent assemblage of nannofossils suggestive of hemipelagic 377

sedimentation and not the mixed assemblage of early Neogene and Paleogene forms found 378

within the MTC, as might be associated with a reworked deposit. We use this noticeable change 379

in nannofossil assemblage as a criteria for defining the top of the MTC. In this study we define 380

both a sedimentary and biostratigraphic top from the core, as well as the top inferred from the 381

strong reflector in the seismic image, typically associated with massive carbonate beds. The 382

sedimentary top of the deposit marks the transition from sediment that is clearly slumped or 383

tilted in the core and appears to have been affected by syn-sedimentary deformation (Figs. 3 and 384

4) while the biostratigraphic top represents the transition from reworked into pristine nannofossil 385

assemblages. The difference in depth, ~35 m, is significant and could represent continued 386

slumping and reworking of young sediments after the initial emplacement of the main MTC 387

bodies. 388

17

The base of the complex is easily established in both drilling sites, being marked by the 389

presence of carbonate breccias immediately overlying fine-grained sediments (Figs. 3 and 5). 390

The depth of this contact is 1101.65 and 1054.1 mbsf at Sites U1456 and U1457, respectively. A 391

key observation is that in the thicker Site U1456 section there is a 20-m-thick interval in which 392

normal hemipelagic sedimentation was briefly reestablished, based on the lack of reworking in 393

the nannofossil assemblages. This spans from around 956 to 935 mbsf (Figs. 3 and 6). This 394

shows that the MTC must have been emplaced in at least two phases separated by a pause, 395

despite the fact that this is not apparent in the seismic image. What is surprising is that the top of 396

this hemipelagic hiatus in mass wasting is not marked by a fresh influx of clearly reworked 397

brecciated carbonate material. Much of the hemipelagic interval comprises massive or parallel-398

laminated mudstones with a couple of medium-bedded to massive sandstones representing less 399

than 10% of the section (Fig. 6A). This is only moderately different from the material which lies 400

above the hemipelagic layer that is characterized by mudstones interbedded with thin beds of 401

siltstone. Above the hemipelagic layer, however, there is clear evidence for slump folding, tilted 402

bedding and microfaulting, which testifies to the redeposited character of these sequences, as 403

well as the mixed nannofossil assemblage. It is only in the somewhat shallower part of the 404

section at Site U1456 there is evidence for a fresh influx of very coarse redeposited carbonate 405

debris flow material, above 874.2 mbsf (Fig. 3A). 406

At both sites, the topmost part of the deposit largely comprises fine-grained, bioturbated 407

claystones and clay-rich siltstones that are otherwise hard to distinguish from the background 408

deposits of the Indus submarine fan, especially when they are not deformed. Tilted bedding is 409

suggestive of deformation but might be interpreted as being coring related. The presence of 410

18

slump folds close to the sedimentary top of each drilled section is, however, more conclusive in 411

demonstrating continued mass wasting above the coarser grained basal units. 412

413

SEDIMENTARY FACIES 414

The sedimentary facies within the MTC were determined on the basis of core 415

descriptions and, in particular, the analysis of sedimentary structures that give clues to the 416

depositional processes that were operating during emplacement. We here describe the major 417

sediment types and provide interpretations of the depositional mechanisms. These are 418

summarized in Figure 3. 419

420

Limestones 421

Short intervals of the MTC comprise coherent sections of fine-grained limestone that 422

show little evidence for the action of high energy reworking depositional processes. Limestones 423

are found at Site U1456 within the lower part of the section around 1050 mbsf depth (Fig. 3). 424

The limestones are typically massive and generally fine-grained micrite with moderate amounts 425

of clay that give them an off-white color. Heavily bioturbated sediment with vertical Zoophycos 426

trace fossils are typical of sedimentation in moderately deep water, often close to the shelf edge 427

(Fig. 7A)(Ekdale et al., 1984; Seilacher, 1967). Figure 7B shows a massive micritic limestone 428

with some evidence of bioturbation, but which indicates minor recrystallization along stylolites, 429

highlighted by thin clay-rich partings. Neither deposit contains indication of strong current 430

activity, such as ripples or laminations, or even a well sorted granular texture, but rather 431

sedimentation in a low energy carbonate-rich environment probably below storm-wave base 432

(<40 m)(Peters and Loss, 2012). Short intervals of limestone are also found at Site U1457 very 433

19

close to the base of the MTC ~1050 mbsf. These are granular and porous and may be the product 434

of higher energy sedimentation in relatively shallow water depths (<30 m). Again, the limestones 435

are tan-colored rather than being pure white that is indicative of a modest clay content. Given the 436

modern significant water depth (3523 m at Site U1457) we propose that these limestones 437

represent coherent blocks of relatively shallow water material that were emplaced as part of the 438

brecciated units near the base of the MTC. 439

440

Carbonate Breccia Debrites 441

The vast majority of the carbonate sediment in the MTC are breccia clasts found mostly 442

in the bottom part of the deposit at Site U1456 (970–1101 mbsf), with further yet more limited 443

clasts in the upper part of the MTC at the same site. They are also found immediately above the 444

base of the MTC at Site U1457 (Fig. 3). These breccias are thick-bedded, ranging close to 20 m 445

thick for individual beds separated by finer grained units. At Site U1456 there are multiple such 446

breccia units, stacked on top of each other, that are preferentially developed towards the base of 447

the sequence. The breccias are sometimes overlain by calcarenites (described below) or by 448

mudstones with a sharp boundary between the two lithologies. The breccias are extremely 449

poorly-sorted and the individual clasts are angular to sub-angular. Clast size ranges up to and 450

greater than the width of the core (>10 cm). There is usually no trend towards fining or 451

coarsening upwards within individual units, although one coarsening upwards sequence is seen 452

in Section U1456D-43R-1 (860 mbsf). The fabric of the sediment is rarely clast-supported (Fig. 453

8A) but is normally suspended in a dark muddy matrix (Fig. 8B). 454

The limestone clasts are pale tan to bright white with the interior showing a very fine-455

grained or slightly granular sediment classified as micrite or more rarely packstone and 456

20

wackestone (Dunham, 1962). In the part of the section densest in limestone clasts (~1036 mbsf at 457

Site U1457), clasts are seen to indent one another both in core surfaces (Fig. 8A), as well as in 458

microscope thin sections (Fig. 9D). We interpret this as a result of dissolution during diagenesis 459

and burial. 460

The vast majority of the carbonate rocks redeposited in the debris flows appear to have 461

been lithified prior to their resedimentation. In combination with the observation of angular 462

clasts, we see coherent rafts of sediment (>10 cm width) floating within finer grained material 463

(Fig. 8B). There is some evidence that some of the carbonate sediment was not lithified during 464

emplacement because soft sediment folding of the deposits, such as seen in muddy limestones 465

(Fig. 10A) can be observed. However, these deformed deposits only represent a relatively small 466

part of the total sequence. It is clear that brittle deformation is important locally, especially 467

between and within the more coherent carbonate blocks. Slickensides especially testify to rapid 468

brittle deformation of the carbonate rocks during their emplacement (Fig. 8C). Most of the debris 469

flow units are extremely poorly-sorted but sometimes are represented by coarse sandstones 470

devoid of larger clasts (Fig. 8D). In these, larger granular clasts are supported in a muddy 471

sandstone matrix with no clear grading within the unit. 472

Although limestone fragments dominate the debris flows, it is noteworthy that in places 473

there is evidence for reworking of volcanic rocks into the flows (Fig. 10B). These clasts are 474

weathered red-brown and are sub-rounded. The largest single clast was found at 879 mbsf at Site 475

U1456 within a poorly indurated conglomeratic part of the debris flow sequence. The clast is an 476

8-cm-wide fragment of vesicular aphyric basalt that is presumed to be derived by erosion from 477

the Deccan Plateau volcanic sequences exposed across peninsular India. The clasts were likely 478

21

eroded on to and then reworked across the continental shelf because being redeposited in the 479

MTC. 480

The limestone, from which the carbonate clasts were derived, formed as a typical 481

shallow-water deposit in a biologically productive zone mostly starved of clastic sediment input. 482

Original water depths were within the photic zone on the continental shelf or within a back-reef 483

setting (<50 m), with only moderate amounts of current activity, since we see no evidence for 484

strong sorting or high energy deposits such as oolites or grainstones (Dunham, 1962). These 485

original rocks have mostly been broken and reworked as debris flow deposits during the 486

emplacement of the MTC. The muddy matrix has a separate provenance, either from the deep-487

water slope of peninsular India or from the Indus Fan itself, as discussed below. 488

489

Calcarenites 490

Calcarenite is present in each carbonate section, in the form of massive, well-sorted units 491

suggestive of high energy current transport. Beds of calcarenite are several meters thick and 492

generally massive and structureless, although they can develop a sub-horizontal fabric suggestive 493

of current flow. Where the deposits are finer (Fig. 10D), there is a shear-type fabric developed 494

within the calcareous siltstones. In the coarser grained units (Fig. 10C) there is some evidence 495

for internal soft sediment deformation, although generally the units are homogenous and 496

comprise uniform, gray, coarse-grained sandstone. They are well-sorted and clast-supported, 497

with very little muddy matrix, suggestive of a high energy depositional regime. The majority of 498

the clasts are carbonate, although there are a significant number of dark grains of organic carbon 499

origin. These calcarenites often have sharp tops that are interpreted to reflect erosion of the 500

deposit prior to the emplacement of overlying units. Figure 10D shows a calcareous siltstone 501

22

sharply overlain by conglomeratic sandstones deposited as debris flows. Very few sedimentary 502

structures are seen within these deposits, so that we infer sedimentation in an upper flow regime 503

resulting in relatively laminar deposits without any current ripples or finer interbeds. Sediment 504

concentrations are inferred to have been very high during deposition, which terminated rapidly. 505

506

Turbidites and Hemipelagic Mudstones 507

Apart from the carbonate-dominated debris flows, minor turbidite sandstones and 508

dominant siltstones and mudstones make up the largest part of the MTC. These are also 509

interbedded with associated hemipelagic mudstones. In the coarsest sandstones, each turbidite 510

shows a classic fining upward sequence (Fig. 11A), with largest carbonate fragments suspended 511

in a dark clastic mud matrix. Locally, there are sub-horizontal lamination although sedimentary 512

structures are poorly developed, with up-section fining dominating characteristic of these 513

deposits. In the upper parts of the MTC at both sites, muds show lamination and interbedding of 514

modest amounts of muddy silt (Fig. 11B). Elsewhere, the deposits are massive, dark gray 515

mudstones with few sedimentary structures. These contrast with the draping mudstones that 516

overlie the catastrophically emplaced MTC where typical deep-water trace fossil assemblages 517

(i.e., Zoophycos;(Fig. 11C) characterize the hemipelagic sedimentation and eliminate the 518

possibility of large-scale mass wasting. This is in contrast to the muddy upper sections of the 519

MTC itself, where there is evidence for laminar current flow that follows the initial emplacement 520

of the carbonate debris flow deposits at the base of each cycle. In general, the grain sizes are 521

relatively limited, with only few a thin-bedded sandstones and occasional siltstones developed 522

within what is otherwise a dominantly (95%) muddy sequence. Distinguishing muddy sediment 523

23

with the MTC from the hemipelagic interval within Site U1456 is difficult without the help of 524

micropaleontology evidence. 525

Syn-sedimentary deformation within the muddy turbidities include folds, micro-faults, 526

and tilted bedding (Fig. 11D) and are particularly easy to see in well-laminated sequences. Dip of 527

lamina can be high (>50°), indicating significant deformation of the muddy units after 528

sedimentation. In addition to ductile structures, there is evidence for compressional reverse 529

faulting. Significant dips and deformation are evidence for incorporation as part of the MTC 530

rather than the subsequent hemipelagic sedimentation of the Indus Fan, which is only gently 531

inclined like the seafloor or the top of the MTC (~1.2˚ according to Calvès et al. (2015)). 532

533

Micro-Facies 534

Petrographic analysis can be used to help interpret paleoenvironment and depositional 535

mechanisms from facies identified in the cores. Figure 9A shows a silty laminated mudstone 536

from the upper part of the MTC at Site U1457 that is interpreted here as a turbidite deposit. The 537

massive calcarenite beds that overlie debris flow conglomerates are seen to be relatively poorly 538

sorted and matrix supported, at least in places, in thin section (Fig. 9B). Clasts are rarely 539

composed of calcite crystals but are dominated by a variety of finer limestone facies, especially 540

micrite. Aggregates of dolomite crystals are observed (Fig. 9C) and interpreted to represent 541

diagenetic alteration of original calcite via interaction with magnesium-rich waters prior to 542

resedimentation. Their presence is suggestive of redeposition from shallow water areas where 543

this mineral generally forms. 544

There are large numbers of microfossils and their fragments within the breccia limestone 545

clasts. Foraminifers are abundant (Figs. 12A, 12B, 12F). In addition, we also confirm the 546

24

presence of crinoid fragments (Fig. 12D), bryozoans, and rare radiolarians (Fig. 12E). The 547

skeletal assemblage of most limestone clasts is dominated by calcareous red algae and benthic 548

foraminifera (including both miliolids and large rotaliids; Fig. 12C). Rare echinoderms, mollusks 549

and hermatypic coral fragments are also present. Some skeletal grains, originating from a 550

shallow-water environment (coralline algae, large echinoid spines, large benthic foraminifera), 551

also occur within the matrix (Figs. 12H, 12I). The occurrence of what is likely to be Lockhartia, 552

together with the peyssoneliacean red-alga Polystrata alba, suggests that at least part of the 553

eroded limestone was of Paleogene age (Fig. 12C)(Bassi and Nebelsick, 2000; BouDagher-554

Fadel, 2018). The matrix is largely dominated by planktonic foraminifera with minor 555

contribution from small rotaliids (Figs. 12G). 556

These characteristics suggest that the MTC involved both lithified inner platform deposits 557

(the source of limestone fragments) and outer platform deposits still composed of loose grains 558

(the source of the muddy matrix with planktonic foraminifera). 559

560

DEPOSITIONAL MECHANISMS 561

Most sediment within the MTC are either debris flow deposits, well-sorted calcarenites, 562

or dominantly clastic turbiditic siltstones and mudstones. Both phases of the MTC at Site U1456 563

(Fig. 3) show large-scale fining upwards cycles, with a dominance of carbonate debris flows 564

towards the base grading into more siliciclastic turbidite sedimentation towards the top. Smaller, 565

shorter phases of fining upwards cycles are further observed within the two overall fining 566

upwards cycles at Site U1456. For example, the upper part of Phase 1 (Fig. 3), comprises a basal 567

unit from between 999.2 and 984.0 mbsf that is dominated by rafted carbonate sheets and 568

carbonate debris flow material (Figs. 3 and 6B). This interval is likely a second pulse after the 569

25

initial Phase 1 event. Above 984.0 mbsf there is a transition to massive thick-bedded calcarenite 570

with slump folds, although this is truncated sharply at 973 mbsf by mudstones that rapidly 571

transition into the hemipelagic sediment described above (Fig. 6B). This implies that the basal 572

Phase 1 unit, especially at Site U1456 comprises a series of pulses rather than one single gigantic 573

deposit as might have been implied by the seismic data alone (c.f.(Calvès et al., 2015)(Fig. 2). 574

The base of Phase 1 at both sites is characterized by a thick-bedded sequence of debris 575

flow calcareous breccias and rafts of undeformed shallow water carbonate (Fig. 5). These are not 576

surprisingly the thickest such deposits within the entire drilled section. Although Site U1456 is in 577

a more central location within the basin, the oldest debris flow breccia at the base of Phase 1 is 578

thinner in this location than at Site U1457 and transitions more rapidly up into thick-bedded 579

breccia and interbedded calcarenites. Both sections, however, do show an overall fining upward 580

between the base and overlying mudstone units. The initial debris flow sedimentation appears to 581

be ~94 m thick at Site U1456 (1101.6–1007.2 mbsf) and ~48 m thick at Site U1457 (1006.4–582

1054.3 mbsf; Figs. 3 and 5). 583

In general, calcarenites alternate with debris flow conglomerates (Fig. 5A) indicating 584

alternating depositional mechanisms within a single emplacement episode. Individual debris 585

flow events are followed by high energy upper flow regime periods of sedimentation where 586

massive well-sorted calcarenites were deposited before being followed by another debris flow 587

unit. However, presumably all this material was emplaced over a relatively short period of time. 588

The carbonate-dominated debris flows form the initial erosive base of the MTC, followed by 589

mud-dominated turbidite sedimentation and hemipelagic fallout representing the tail of the MTC. 590

At Site U1456 this sequence is then repeated after the hemipelagic break. Soft sediment 591

deformation is commonly seen in the more laminated sections indicative of slumping after 592

26

sedimentation. It seems unlikely that poorly consolidated mudstones and siltstones could have 593

been emplaced hundreds of kilometers in a semi-coherent form, unlike the well-lithified 594

limestone clasts seen close to the base of each section. 595

596

GEOCHEMISTRY 597

Bulk Geochemistry 598

We use a CN-A-K ternary diagram to illustrate major element geochemistry of MTC 599

samples compared to sediments from the Indus Canyon and delta. The sediment from the MTC 600

largely plots within the range of the Indus Canyon and trends towards higher values of Al2O3 601

(Fig. 13A). MTC samples appear to have higher values that trend towards the illite end-members 602

and may be more depleted in biotite and feldspars compared to the delta. This is likely a result of 603

sediment transport, similar to what has been observed in the Indus Canyon (Li et al., 2018). 604

Sediments in the muddy upper part of the MTC at Site U1457 largely plot with low 605

Chemical Index of Alteration (CIA), which is a proxy of the state of weathering of a sediment 606

compared to pristine bedrock (Nesbitt et al., 1980). The muddy upper MTC samples trend more 607

towards the Quaternary Indus Delta field compared to the lower parts of both Phase 1 and Phase 608

2, which show more overlap with western Indian Shelf sediments, largely derived from rivers 609

draining the Deccan Plateau (Kurian et al., 2013). This plot implies that the upper muddy 610

sediments at Site U1457 had a dominant source from the Indus River/Fan and little inputs from 611

western peninsular India. 612

The sediment in the MTC can also be characterized using other major element 613

discrimination diagrams. Figure 13B shows the scheme of Herron (1988) in which the Phase 1 614

and Phase 2 samples largely plot within the Fe shale field, with a few slightly depleted in Fe and 615

27

plotting as shales. Again, we plot these samples along with the western Indian Shelf, Indus 616

Canyon and delta sediments. Samples from the upper muddy top to Phase 1 at Site U11457 form 617

a cluster within the range of the Indus Canyon sediments, suggesting a dominant provenance of 618

reworked Indus material. Comparison with sediment from the western Indian shelf shows a 619

significant difference, with the shelf sediment typically plotting with much higher Fe contents, 620

similar to the lower Phase 1 and 2 sediments. We infer that the bulk of the sediment in the lower 621

MTC comprises mostly Indian margin sediment with muddy top dominated by sediment eroded 622

and redeposited from the Indus Fan. 623

624

Nd and Sr Isotopes 625

We use Sr and Nd isotope values to constrain the provenance of siliciclastic sediment in 626

the MTC. By cross-plotting Nd and Sr isotopic compositions from source regions such as the 627

Deccan Traps, peninsular Indian rivers, Transhimalaya, Karakoram, Greater Himalaya, Kirthar 628

and Sulaiman Ranges, and modern/Quaternary Indus-derived sediment allows the origin of the 629

sediment to be further constrained (Fig. 14). This diagram shows that the MTC samples form a 630

relatively discrete cluster with one exception that has especially positive Nd values that fall 631

within the Deccan and Transhimalayan arrays. When we compare these data with potential 632

sources, it is clear that the bulk of the sediments lie within the isotopic range defined by the 633

Indus submarine fan sediments at the same drilling sites (Clift et al., 2018). This is consistent 634

with the argument that much of this material may be reworked Indus-derived sediment. 635

However, we note that it is impossible to exclude mixing of sediment from the peninsular Tapti 636

or Narmada Rivers. The isotope compositions by themselves do not allow us to quantify the 637

degree of reworking from these sources as they are similar to the Indus. Although the MTC 638

28

samples plot with higher Nd values compared to the Quaternary Indus Canyon, as well as the 639

Kirthar and Sulaiman ranges, such a composition could largely be explained through temporal 640

variation in the Indus River itself (Clift and Blusztajn, 2005; Clift et al., 2018). The one very 641

positive Nd sample is anomalous and plots with even more positive values than the Tapti River. 642

This is strongly suggestive of erosion from peninsular India and is corroborated by the presence 643

of vesicular Deccan Plateau basalt fragments as previously noted. 644

We can look at the stratigraphic variation in isotopic compositions through time at both 645

sites (Fig. 15). In both cases, Nd isotope compositions plot within error of the Quaternary Indus 646

or with slightly more positive Nd values. We note that the most positive Nd values in each 647

borehole are found within the debris flow conglomerate units bearing basaltic clasts at the base 648

of the lower part of the MTC. This is especially true at Site U1456 (Fig. 15A). Variations in 649

87Sr/86Sr also mirror this general evolution. 650

The provenance of the coarse-grained carbonate debris flow deposits is different from 651

those of the finer grained sediments overlying them. The fine-grained sediments may represent 652

recycling of pre-existing fan sediments into the top of the MTC, while the debris flow deposits 653

are more closely associated with mass wasting from the western Indian continental margin. It is 654

possible that some Indus River sediment could have been transported east along the shelf, carried 655

by longshore currents from the river mouth, and deposited offshore Saurashtra before being 656

redeposited as part of the MTC. However, there is no evidence that significant Indus sediment 657

travels farther east than the Rann of Kutch (Khonde et al., 2017; Kurian et al., 2013). The 658

simplest interpretation is that the upper muddy layers of the MTC represent entrained and 659

reworked Indus Fan material. 660

661

29

Heavy Mineral Analysis 662

The heavy-mineral assemblages help to constrain the source area of the MTC. The 663

concentration of heavy minerals in all samples is very low suggesting a strong depletion due to 664

intrastratal dissolution of unstable silicates (Garzanti, 2017). Consequently, a relative enrichment 665

of ultrastable minerals is observed (ZTR index of Hubert (1962)). The two samples (U1456E-666

15R-1W, 61-63 cm and U146E-17R-4W, 131-133 cm), analyzed from the carbonate breccia 667

present extremely low HMC (0.04–0.05%) with common augitic clinopyroxene (~6%) and rare 668

spinel (2–3%). The minerals also show corroded surficial textures, indicating a strong diagenetic 669

overprinting (Ando et al., 2012). A similar fingerprint is detected in Sample U1456E-7R-1, 80-670

82 cm where green and brown augite are abundant (48%). In all these samples, there are 671

common garnets associated either with apatite, titanite, epidote, zircon, tourmaline, and 672

metamorphic Ca-amphiboles, potentially derived from recycled sediments from the Himalaya-673

derived Indus Fan turbidites eroded by the MTC. Notwithstanding diagenetic dissolution, the 674

highly unstable augitic clinopyroxene (volcanic origin) always dominates over metamorphic 675

amphiboles, suggesting a sizable contribution to the MTC from the Indian passive margin, and 676

especially from Deccan Plateau basaltic lavas. Sample U1456E-4R-1W, 110-111 cm is a 677

calcarenite within which hydraulic sorting and high-energy currents preferentially selected the 678

available heavy minerals suite derived from the MTC, concentrating platy heavy minerals such 679

as chloritoid, Ca-amphiboles and tourmaline (lighter). The sample is partially depleted in denser 680

garnet. This assemblage is completed with the presence of abundant apatite, common titanite, 681

epidote and spinel with trace of kyanite, andalusite and staurolite. 682

Sample U1457C-88R-4W, 58-60 cm was deposited far from the Indian Passive margin 683

and the mineralogy reflects a dominant contribution from recycled minerals derived from the 684

30

erosion and re-deposition of the Indus Fan turbidites. The tHMC is very low (0.08%), and 685

mineralogy is dominated by abundant epidote and garnet with common apatite and titanite. Ca-686

amphiboles dominate over clinopyroxenes, with a ratio 8:1, pointing to a major contribution 687

from the Indus River and the Himalaya in this sample. The assemblage also includes tourmaline, 688

zircon, chloritoid, Cr-spinel and trace of and kyanite, staurolite and andalusite. 689

The modern Tapti River was analyzed close to its mouth. The sample contains a very rich 690

assemblage of heavy minerals (tHMC 17%) with dominant augitic clinopyroxenes (92%) and 691

subordinate amount of metamorphic heavy-mineral, Ca-amphiboles, epidote, garnet and 692

sillimanite. This mineralogical signature differs from the observed suite of orogenic heavy 693

minerals observed in the modern Indus River and his delta (Garzanti et al., 2005). 694

The heavy mineral assemblage in the MTC and the very low concentration of heavy 695

minerals points to different sources for the siliciclastic sediments, i.e., partially derived axially 696

from the Himalayas via the Indus River (especially at Site U1457C) and partially derived 697

transversally from the Indian peninsula (especially at Site U1456). 698

699

Zircon U-Pb Ages 700

To further constrain provenance, we compare detrital zircon U-Pb ages with existing data 701

from the Indus river mouth (Clift et al., 2004), Indus Fan turbidites above and below the MTC 702

(Clift et al., 2018), and with bedrock data from potential sources in the river catchment (Fig. 703

16)(DeCelles et al., 2000; Gehrels et al., 2011). Although the zircon ages from source bedrock 704

overlap with each other, each source regions demonstrates strong preferential age spectra that 705

can be used to discriminate between them. Zircons from Nanga Parbat, Kohistan, the 706

Transhimalaya, and the Karakoram generally have younger ages (<300 Ma) than those from the 707

31

Himalayan ranges (Alizai et al., 2011)(Fig. 16). Both the Greater and Tethyan Himalaya have U-708

Pb age peaks at 300–750 Ma and 750–1250 Ma, with older ages at ~1850 Ma characterizing the 709

Lesser Himalaya 710

The volume of sample available for U-Pb dating from Core U1457C-7R (the only suitable 711

sediment seen in the MTC) was extremely limited such that only 51 grains yielded concordant 712

ages, which is somewhat lower than the 113 minima suggested by Vermeesch (2004) for a 713

sample with complex provenance. Nonetheless, some inferences concerning provenance can be 714

made. What is clear is that young ages dominate with 17 grains dated at less than 100 Ma (Fig. 715

16). The age spectrum bears most similarity with Indus Fan turbidites dated at 7.8, 8.3, and 15.6 716

Ma, but all are in contrast with the ages from the modern river. The match between these young 717

grains and sources in the Karakoram and Kohistan argue for the sand to be an Indus-derived 718

sediment and not from sediment transported from the Indian peninsula where zircon ages are 719

Paleozoic or typically much older. This conclusion is consistent with the Nd and Sr isotope data 720

from the upper parts of the MTC. The analyzed sandstone was sampled below the 721

sediment/structurally defined top of the MTC but above the carbonate-dominated debris flow 722

facies at the base of the complex, i.e., within the muddy but slumped top of the MTC. This 723

implies that the upper parts of the MTC are Indus Fan sediments entrained in the tail of the MTC 724

during emplacement. 725

726

Clay Mineralogy 727

The clay mineral assemblages within the MTC can be used to assess provenance by 728

semiquantitative analysis and comparison with existing data from other sources. When plotted on 729

the ternary diagram of (illite+chlorite), kaolinite, and smectite (Fig. 17) there is significant 730

32

overlap between the new MTC data and other Arabian Sea sediments (Rao and Rao, 1995). In 731

general, the MTC clays are low in kaolinite and form an array between the smectite and 732

(illite+chlorite) end members. In this respect, they show a similar character to sediments from the 733

Indus fan and have significant overlap with Quaternary clays from the Indus Canyon (Li, 2018). 734

Samples from Phase 1 of the MTC have very high smectite contents, similar to the Paleocene 735

sediments overlying basement at Site U1457, suggestive of a volcanic source. They are close to 736

sediments recovered from the inner shelf offshore Saurashtra and from the Gulf of Cambay. 737

Phase 2 sediments and the hemipelagic layer are slightly less smectite rich but overlap with the 738

Holocene Indus Shelf, as well as some modern Indian Shelf sediments. We note that the bulk of 739

the muddy upper Phase 1 sediments plot with higher (illite+chlorite) values and they also tend to 740

have slightly higher kaolinite compared with analyses of sediments from the Indus floodplains 741

(Alizai et al., 2012). These sediments are similar to the assemblage recognized from the outer 742

Saurashtra margin (Rao and Rao, 1995) and are similar to many clay assemblages within Indus 743

Fan turbidite sequences. Overall, the MTC deposits have lower kaolinite compared with most 744

Western Indian shelf deposits but some samples plot closely to the shelf. It is also noteworthy 745

that the MTC assemblages generally show lower (illite+chlorite) compared with many of the 746

Miocene-Recent Indus submarine fan deposits, which likely indicates a mixed provenance of 747

Indus and Indian peninsular sediment. However, because illite and chlorite are the product of 748

physical weathering rather than chemical weathering their relatively high contribution to the 749

MTC could also indicate reduced chemical weathering of fan sources since MTC emplacement. 750

These data are consistent with a dominant recycling of Indus Fan deposits in the upper muddy 751

parts of the MTC, but with greater involvement of clays derived from the Western Indian margin 752

33

in the lower part, especially in Phase 1. The similarity with modern nearshore sediments offshore 753

Saurashtra and Cambay is consistent with an origin in this part of the margin. 754

Clay mineralogy shows significant variation with depth (Fig. 15). At Site U1456 the 755

carbonate-rich part of the section shows particularly high smectite contents and relatively low 756

(illite+chlorite) values. Smectite only becomes less abundant than these two physically 757

weathered clays above the upper Phase 2 carbonate debris flow unit. At Site U1457 the 758

carbonate-rich part of the section similarly is smectite-rich, but immediately above this level the 759

sediments become dominated by an (illite+chlorite) assemblage similar to the Indus Fan. It is 760

noteworthy that the Paleocene sediments beneath the MTC at Site U1457 are ~100% smectite, 761

possibly reflecting chemical weathering of the underlying basaltic basement. Clay mineralogy 762

supports the Nd and Sr isotope compositions in showing a characteristic difference between the 763

carbonate-dominated sections that indicate similarity to the western Indian margin, whereas the 764

mudstone dominated sequences further upsection in the MTC are most similar to compositions 765

associated with the Indus Fan. 766

767

SEDIMENT BUDGET 768

To assess the potential of sediment delivery rates and margin oversteepening as triggering 769

mechanisms of the MTC, a sediment budget from the western Indian margin was generated using 770

standard two-dimensional backstripping methods from seismic profile data (Clift, 2006; Kusznir 771

et al., 1995). This was to primarily test the hypothesis that the rapid accumulation of sediment on 772

the continental margin resulted in an unstable stratigraphy that was then more liable to mass 773

wasting events like the Nataraja MTC. There is strong evidence that the Western Indian 774

continental margin is gravitationally unstable as a result of the large-scale compressional thrusts 775

34

seen in seismic profiles towards the base of the continental slope seen between the Saurashtra 776

shelf and Bombay High (Fig. 1)(Calvès et al., 2015; Nair and Pandey, 2018). These features are 777

often associated with slopes prone to gravitational collapse, which in this region, has yet to 778

manifest in the dramatic fashion of the Nataraja MTC. In order to estimate the mass flux of the 779

margin, we use the cross-margin seismic reflection profile of Nair and Pandey (2018)(Figs. 1 and 780

18). Their northernmost profile lies immediately south of the scarp region identified by Calvès et 781

al. (2015) and which we consider to be potentially representative of the sedimentation in the 782

source regions of the MTC prior to its redeposition. For the purpose of this study, we use the age 783

control provided by Nair and Pandey (2018), at least for the continental shelf and slope areas 784

(Fig. 18A). West of the toe of the slope sedimentation is linked to the Indus Fan and may not be 785

representative of the mass flux to the Saurashtra Shelf. Figure 18A shows the interpretation of 786

Nair and Pandey (2018) with a conversion from their seismic travel time scale to depth made on 787

the basis of multichannel seismic stacking velocities derived from the Indus shelf, as used by 788

Clift et al. (2002)(Table 6). We do this because of the absence of such data from the Saurashtra 789

region itself. We prefer to use velocity data from the Indus continental shelf rather than from the 790

deep basin because as the sediment thicknesses are much greater under the continental shelf, they 791

are more comparable to those seen offshore the Indus River mouth. Based on the lateral 792

variability in velocities seen on the Indus Shelf, we estimate that this conversion may introduce 793

uncertainties as high as ±20% (Clift, 2006). Stratigraphic ages are then assigned numerical ages 794

based on the timescale of Gradstein et al. (2012). 795

The depth-converted line was then backstripped using standard decompaction methods 796

(Kusznir et al., 1995; Sclater and Christie, 1980). This was done to restore each dated sediment 797

layer to its original thickness prior to burial. Knowledge of the sediment type is important to this 798

35

calculation because shales experience much greater loss of porosity during burial than do 799

sandstones (Sclater and Christie, 1980), and in this case, we used lithological data from Wandrey 800

(2004) and Rao and Talukdar (1980). These studies show a mixed Cenozoic sequence dominated 801

by silty muds and carbonates offshore Saurashtra. The decompaction process involves 802

accounting for the loss of porosity of the sediment during burial, which would otherwise result in 803

an underestimation of deposited volumes for the older, deeper buried sediment packages. After 804

the original, uncompacted volume of sediment in each dated interval has been determined, the 805

mass of rock delivered during that time period can be calculated. Errors in lithology and 806

compaction history are much smaller than the time-depth conversion and rarely exceed 5%. 807

In this study two-dimensional decompaction was calculated using the program Flex-808

Decomp™ (Kusznir et al., 1995). It must be assumed that the analyzed profile is representative 809

of the total mass flux to the margin since rifting of the Arabian Sea ~66 Ma (Bhattacharya et al., 810

1994). Because we only have one profile close to the area of mass wasting, and no estimate of 811

the total sediment mass offshore Saurashtra, it is not possible to make a volume calculation. 812

However, the two-dimensional budget does at least allow us to estimate the volumes of sediment 813

delivered per kilometer of margin close to the source of the MTC. Our results show a clear trend 814

to increasing mass flux after 26 Ma (Fig. 18B), with a peak between 16 and 11 Ma. Because the 815

resolution of the budget is constrained by the presence of the dated horizons, it is not possible to 816

accurately say when the peak sediment flux was achieved, but this analysis confirms that the 817

Middle Miocene was a time of rapid sedimentation offshore Saurashtra, a pattern that it shares 818

with many other Asian delta systems. As a result, it seems likely that the pulse was caused by 819

faster erosion driven by heavy summer monsoon rains (Clift, 2006). We suggest that much of the 820

gravitational instability on the western Indian margin was caused by rapid sedimentation in the 821

36

Middle Miocene causing oversteepening of the shelf edge, comprising large thicknesses of 822

sediment liable to incomplete dewatering during burial. The reducing sedimentation rates after 823

11 Ma may explain why a second such slide has not been emplaced in this part of the margin. 824

825

SEISMICITY 826

As well as an over-steepened continental margin caused by increased sediment flux, we 827

investigate the possible triggering of the MTC as a result of seismic activities that are often 828

implicated in the emplacement of large mass wasting complexes (Kastens, 1984). Figure 1 shows 829

the location of earthquakes greater than 4.5 magnitude since 1960 in the vicinity of the source 830

region for the MTC. There is some seismicity related to the plate boundary west of the Indus 831

delta and there are small amounts of activity in the Saurashtra Peninsula itself, immediately 832

opposite the scar in the continental shelf. It is apparent that the greatest concentration of seismic 833

activity is however around the Rann of Kutch, where historic intraplate events up to 7.7 834

magnitude have been recorded (Bilham, 1999). This activity reflects reactivation of earlier rift-835

related faults due to compression linked to the India-Eurasia collision (Bilham et al., 2003; 836

Biswas, 2005). This part of the Indian plate is a weak zone and may well have been active as a 837

seismic hotspot for significant periods of time. We suggest that it is the relative proximity of the 838

Saurashtra margin to this tectonic feature (<300 km) which may have initiated the mass wasting 839

in that region, rather than further south along the margin where sediment flux was also high. 840

841

SYNTHESIS AND CONCLUSIONS 842

This study, made possible through drilling, reveals for the first time the internal structure 843

and origin of the Nataraja MTC, and extends our understanding based on the earlier seismic 844

37

surveying of the deposit. At Site U1456, there is clear evidence that the MTC was emplaced in 845

two major phases separated by a significant break (Fig. 19). Even the larger, earlier Phase 1 can 846

be broken down into at least two stages, indicative of pulsed emplacement. The basal part of 847

each drilled section of the complex comprises debris flow carbonate breccias and larger rafts of 848

shallow water limestone, which can be traced back to collapse of the carbonate edge of the 849

continental shelf offshore Saurashtra. The MTC is emplaced as a number of fining upward 850

sequences with debris flow breccias, overlain by well sorted, coarse calcarenite deposited by 851

high velocity currents following in the wake of the initial mass wasting landslide. These are 852

overlain by muddy and turbiditic deposits, which are increasingly siliciclastic in character. At 853

Site U1457, only a thinner section of the earlier Phase 1 appears to be preserved, but a second 854

Phase 2 is apparent at Site U1456. Again, there was an emplacement of carbonate-rich debris 855

flows, although these were preceded and followed by muddy turbidite deposits, largely reworked 856

from pre-existing sediments of the Indus Fan. The top of each drilled sequence shows a 857

separation between sediment where the biostratigraphy is mixed and where slumping continues 858

to occur in the aftermath of the original depositional event. 859

Nd and Sr isotopic data, together with heavy-mineral assemblages, show that the 860

siliciclastic fraction of the deposit is associated with the western Indian continental margin, at 861

least in the debris flow part of the deposits although the overlying muddy turbidite units share the 862

same characteristics as the Indus submarine fan and suggest entrainment of sediment already 863

deposited in Laxmi Basin in the wake of the carbonate-rich debris flows that formed the MTC in 864

the first place. Limited zircon data at Site U1457 also show the clear signature of the Indus 865

River, although this applies only to the muddy units overlying the carbonate debris flows. We 866

envisage that enhanced sediment delivery to the western Indian continental margin driven by 867

38

strong monsoon during the middle Miocene resulted in an oversteepened continental margin that 868

was in a gravitationally unstable state. Exactly what triggered the collapse is not clear, but may 869

well be related to seismic activity in the nearby Rann of Kutch where large earthquakes continue 870

to the present day. Compressional deformation structures in the western Indian continental 871

margin south of Saurashtra suggest that this region too is in a compressional and potentially 872

unstable situation. However, decreasing sediment flux to the continental margin since the middle 873

Miocene has lessened the instability of the continental slope and reduced the chance of mass 874

wasting, especially further south away from potential seismic triggers. The western Indian 875

margin, however, has also experienced the increasing sedimentation rates linked to the onset of 876

northern hemisphere glaciation and so the potential for significant geohazard still exists. 877

Nonetheless, the fact that there has been no similar large event since 10.8 Ma does argue for this 878

being relatively low risk at the present time. 879

880

Acknowledgments 881

This work was made possible by samples given by International Ocean Discovery 882

Program. Project funding came from the Charles T. McCord Jr Chair in Petroleum Geology at 883

LSU. We thank Alan Roberts and Nick Kusznir for use of FlexDecomp software. 884

885

886

39

Figure Captions 887

Figure 1. A) Shaded topographic and bathymetric map of the Arabian Sea showing the location 888

of the core sites discussed in this study (yellow dots). Base map from GeoMapApp. Dashed 889

yellow lines show proposed continent-ocean boundaries. Dashed white lines show oceanic 890

transform faults. Numbered red circles indicate existing scientific boreholes from Deep Sea 891

Drilling Project (DSDP) and Ocean Drilling Program (ODP). Pink squares show major cities. 892

Magnetic anomalies (thin gray numbered lines) are from Miles et al. (1993). Green-filled circles 893

show earthquakes >4.5 magnitude since 1960 recorded by US Geological Survey. B) Close-up 894

map of Laxmi Basin showing the precise location of the drill sites. A pink dashed line shows the 895

extent of the Nataraja MTC (Calvès et al., 2015). Light blue lines show locations of seismic 896

profiles shown in Figure 2. 897

898

Figure 2. Seismic profiles of the core sites (left) with interpretation (right) showing the mass-899

transport complex in the immediate vicinity of (A) IODP Site U1456 and (B) IODP Site U1457. 900

Modified from Pandey et al. (2016c). See Figure 1 for locations of lines. 901

902

Figure 3. Summary stratigraphic columns showing the lithologies and interpreted facies of the 903

mass-transport complex at (A) IODP Site U1456 and (B) IODP Site U1457. Black shading in 904

second column indicates recovery, with white showing lost section. mbsf = meters below 905

seafloor. 906

907

Figure 4. (A) Sedimentary log showing the top of the deposit, U1456D-33R to U1456D-42R; (B) 908

Sedimentary log showing the top of the deposit, U1457C-69R to U1457C-78R. Black shading in 909

40

second column indicates recovery, with white showing lost section. mbsf = meters below 910

seafloor. 911

912

Figure 5. (A) Sedimentary log showing the bottom of the MTC, U1456E-16R to U1456E-19R; 913

(B) Sedimentary log showing bottom of the MTC, U1457C-86R to U1457C-92R. Lithological 914

patterns and sedimentary structures same as Figure 4. Black shading in second column indicates 915

recovery, with white showing lost section. mbsf = meters below seafloor. 916

917

Figure 6. (A) Sedimentary log showing the deposit above and within the hemipelagic layer, 918

U1456D-50R to U1456D- 53R. As shown, soft sediment deformation occurs until pelagic layer 919

begins; (B) Sedimentary log showing the second pulse of carbonate debris flow material, 920

U1456D-56R to U1456D-61R. Lithological patterns and sedimentary structures same as Figure 921

4. Black shading in second column indicates recovery, with white showing lost section. mbsf = 922

meters below seafloor. 923

924

Figure 7. (A) Limestone with burrows (20 cm long), U1456E-12R-1, 42-47 cm (1045 mbsf); (B) 925

Stylolite in limestone, U1456E-10R-3, 30-40 cm (1030 mbsf). Vertical scale is in cm below the 926

section top. See locations on Figure 3. 927

928

Figure 8. (A) Coarse carbonate breccia with mudstone matrix, U1457C-90R-2, 75-83 cm (1036 929

mbsf); (B) Debris flow conglomerate with faulted mudstone raft (larger faults shown with white 930

lines), U1456E-9R-4, 78-88 cm (1021 mbsf); (C) Core photograph of slickensides on a fault 931

within silty claystone, U1456E-9R-4, 37-51 cm (1021 mbsf), (D) Coarse sandy, calcarenite, 932

41

U1457C-88R-5, 38-48 cm (1022 mbsf). Vertical scale is in cm below the section top. See 933

location on Figure 3. 934

935

Figure 9. Thin section plane polarized photomicrographs of (A) Laminated sandy siltstone with 936

quartz grains, U1457C-85R-1, 22-26 cm (997 mbsf). Note the finer muddy center of the image 937

and the poorly sorted silt interbeds on either side with dominant quartz clasts; (B) Calcarenite, 938

U1456D- 60R-1, 13-17 cm (1006 mbsf); (C) Euhedral calcite/dolomite crystals within larger 939

grain, U1456E-15R-1, 12-16 cm (1073 mbsf); (D) Suture grain contact of carbonate clasts in 940

breccia, U1456D-45R-4-52-57 cm (870 mbsf). See location on Figure 3. 941

942

Figure 10. (A) Slump folded calcareous siltstone, U1456D-58R-2, 43-53 cm (989 mbsf); (B) 943

Deccan vesicular basalt clast, U1456D-46R-1, 16-25 cm (879 mbsf); (C) Massive calcarenite 944

with ductile folded layer U1456D-41R-3A, 114-124 cm (841 mbsf); (D) Sharp contact between 945

calcarenite and calcareous siltstone, U1457C-88R-7, 61-70 cm (1025 mbsf). Vertical scale is in 946

cm below the section top. See location on Figure 3. 947

948

Figure 11. (A) Sandy siltstone showing gradual normal grading, U1457C-71R-3, 101-115 cm 949

(865 mbsf) (B) Tilted, laminated turbidite deposit U1457C- 73R-2, 140-148 cm (881 mbsf); (C) 950

Mudstone with Zoophycos burrows (one outlined for clarity), U1457C-68R-1, 42-52 cm (832 951

mbsf); (D) Steeply dipping laminated mudstone showing reverse faulting, U1456D-46R-3A, 952

139-148 cm (883 mbsf). Vertical scale is in cm below the section top. See location on Figure 3 953

954

42

Figure 12. Thin section plane polarized photomicrographs of (A) Uniserial benthic foraminifer in 955

breccia clast, U1456E-15R-1, 12-16 cm (1072 mbsf); (B) Siltstone with planktonic foraminifers, 956

U1456D-58R-2, 40-44 cm (989 mbsf); (C) Limestone clast with a specimen of Lockhartia, 957

U1456E-17R-4, 131-133 cm (1086 mbsf); (D) Echinoderm spine in carbonate clast, U1456D-958

61R-2, 44-48 cm (1017 mbsf); (E) Foraminifer fragments in siltstone, U1456D-58R-2, 40-44 cm 959

(989 mbsf); (F) Planktic foraminifers and bioclasts in carbonate breccia, U1457C-90R-1-6-10 cm 960

(1034 mbsf). G) Planktonic foraminifer, U456E-7R-1, 80-82 cm (999 mbsf); H) Fragments of 961

coralline algae (white arrows) included in the planktonic-foraminifer-dominated matrix; Plk = 962

planktonic foraminifer, U1457C-88R-4, 58-60 cm (1021 mbsf); I) Orthophragminid fragment 963

(white arrow) included in the planktonic-foraminifer-dominated-matrix; Dl = dolomite crystal, 964

U1457C-88R-4, 58-60 cm (1021 mbsf). See locations on Figure 3. 965

966

Figure 13. (A) Geochemical signature of the analyzed samples illustrated by a CN-A-K ternary 967

diagram (Fedo et al., 1995). CN denotes the mole weight of Na2O and CaO* (CaO* represent the 968

CaO associated with silicate, excluding all the carbonate (Singh et al., 2005)). A and K indicate 969

the content of Al2O3 and K2O respectively. CIA values are calculated and shown on the left side, 970

with values correlated with on the CN-A-K ternary. Samples from the delta have the lowest CIA 971

values and indicate high contents of CaO and Na2O and plagioclase. Abbreviations: sm 972

(smectite), pl (plagioclase), ksp (K-feldspar), il (illite), m (muscovite). B) Geochemical 973

classification of sediments from the Indus delta (Clift et al., 2010), Indus Canyon (Li et al., 2018) 974

and western Indian Peninsular shelf north of Goa (Kurian et al., 2013) following the scheme of 975

Herron (1988). Phase 1 and Phase 2 sediments, together with the hemipelagic drape are the 976

materials of the MTC. 977

43

978

Figure 14. Cross plot of Sr versus Nd isotope data from the MTC, adjacent drill sites, major 979

source regions onshore, and modern Mahi, Tapti, and Narmada River sediments (Goswami et al., 980

2012). Kirthar and Sulaiman data is from Zhuang et al. (2015). Deccan Plateau data are from 981

GEOROC compilation (http://georoc.mpch-mainz.gwdg.de/georoc/). Transhimalaya data are 982

from Rolland et al. (2002), Singh et al. (2002), and Khan et al. (1997). Greater Himalayan data 983

are from Ahmad et al.(2000), Deniel et al. (1987), Inger et al. (1993) and Parrish and Hodges 984

(1996). Karakoram data are from Crawford and Searle (1992) and Schärer et al. (1990), 985

986

Figure 15. Downhole plots of Nd and Sr isotope compositions and clay mineralogy of 987

siliciclastic sediments from IODP sites (A) U1456 and (B) U1457. Gray vertical bar shows 988

compositional range of Quaternary sediments in the Indus Delta (Clift et al., 2010; Clift et al., 989

2008b), as well as modern Tapti and Narmada River sediments (Goswami et al., 2012). Deccan 990

Plateau volcanic rocks plot outside this range at more positive Nd values and lower 87Sr/86Sr 991

values. Nd and Sr isotope analyses include errors recently suggested by Jonell et al. (2018) for 992

bulk sediment compositions. Error bars encompass the total expected error for any bulk sample 993

as a result of variable grain size and mineralogy, and analytical error contributed during sample 994

preparation, homogenization, and analysis. 995

996

Figure 16. Kernel density estimate (KDE) plots for detrital zircon U-Pb ages for the Nataraja 997

MTC compared to major source terrains in the western Himalayas, from the compilation of 998

Alizai et al. (2011), as well as a modern sand from the river mouth (Clift et al., 2004) and select 999

Indus Fan turbidites also from IODP Sites U1456 and U1457 (Clift et al., 2018). Deccan zircons 1000

44

at ~65 Ma would plot within the Karakoram-Kohistan range but the inset box at the top shows 1001

that grains <100 Ma from the MTC do not cluster at this age and are better match to sources in 1002

the Indus suture zone. Data from the Tethyan, Greater and Lesser Himalaya are compiled from 1003

DeCelles et al. (2004). Karakoram data is from Le Fort et al. (1983), Parrish and Tirrul (1989), 1004

Schärer et al. (1990), Fraser et al. (2001) and Ravikant et al. (2009). Nanga Parbat data is from 1005

Zeitler and Chamberlain (1991) and Zeitler et al. (1993), Transhimalayan data is from Honegger 1006

et al. (1982), Schärer et al. (1984), Krol et al. (1996), Weinberg and Dunlap (2000), Zeilinger et 1007

al. (2001), Dunlap and Wysoczanski (2002), (Singh et al., 2007), and Ravikant et al. (2009). 1008

1009

Figure 17. Ternary diagram of clay minerals from IODP Site U1456 and U1457 indicates a clay 1010

mineral assemblage consisting mostly of smectite, chlorite and illite. Clay mineral data from 1011

source regions are plotted to compare their clay assemblages. Data from western Indian shelf 1012

modern sediments are from Rao and Rao (1995), Indus canyon data is from Li et al. (2018), 1013

Indus flood plain and delta data is from Alizai et al. (2012), and Indus Fan data is from Peng 1014

Zhou (unpublished). 1015

1016

Figure 18. (A) Interpretation of the depth-converted seismic section of the western Indian 1017

continental shelf immediately to the south of the source region for the Nataraja Slide based on 1018

the seismic profile of Nair and Pandey (2018) and using the seismic velocities shown in Table 5; 1019

and (B) A calculated sediment budget for the Indian shelf derived from two-dimensional 1020

sediment backstripping of this profile derived from FlexDecomp software. 1021

1022

45

Figure 19. Schematic cartoon illustrating the over-steepened Indian margin (A), the first phase 1023

of emplacement of the Nataraja MTC (B) separated by a short time of quiescence with 1024

hemipelagic sedimentation (C) from the second smaller phase of emplacement (D). 1025

1026

Table 1. Bulk sediment geochemistry analyzed by ICP-ES. 1027

1028

Table 2. Neodymium and strontium isotope data. 1029

1030

Table 3. Heavy mineral data. HM = heavy minerals; tHM = transparent heavy minerals. The 1031

ZTR index is the sum of zircon, tourmaline and rutile over total transparent heavy minerals 1032

(Hubert, 1962) and is classically used to estimate the mineralogical durability of the assemblage 1033

(i.e., the extent of recycling and/or intrastratal dissolution). 1034

1035

Table 4. U-Pb zircon data for sample U1456C-71R-1, 110 cm. 1036

1037

Table 5. Quantitative estimates of major clay mineral assemblages. 1038

1039

Table 6. Seismic interval velocities for the main stratigraphic units interpreted by Nair and 1040

Pandey (2018) used to depth convert the seismic profile before backstripping. 1041

1042 1043 1044

46

References 1045

Ahmad, T., Harris, N., Bickle, M., Chapman, H., Bunbury, J., and Prince, C., 2000, Isotopic 1046 constraints on the structural relationships between the Lesser Himalayan Series and the 1047 High Himalayan Crystalline Series, Garhwal Himalaya: Geological Society of America 1048 Bulletin, v. 112, no. 3, p. 467-477. 1049

Alizai, A., Carter, A., Clift, P. D., VanLaningham, S., Williams, J. C., and Kumar, R., 2011, 1050 Sediment provenance, reworking and transport processes in the Indus River by U-Pb 1051 dating of detrital zircon grains: Global and Planetary Change, v. 76, p. 33-55. 1052 doi:10.1016/j.gloplacha.2010.11.008. 1053

Alizai, A., Hillier, S., Clift, P. D., and Giosan, L., 2012, Clay mineral variations in Holocene 1054 terrestrial sediments from the Indus Basin; a response to SW Asian Monsoon variability: 1055 Quaternary Research, v. 77, no. 3, p. 368–381. doi:10.1016/j.yqres.2012.01.008. 1056

Andò, S., and Garzanti, E., 2014, Raman spectroscopy in heavy-mineral studies. In Sediment 1057 Provenance Studies, in Scott, R. A., Smyth, H. R., Morton, A. C., and Richardson, N., 1058 eds., Sediment Provenance Studies in Hydrocarbon Exploration and Production, Volume 1059 386: London, Geological Society, p. 395–412. 1060

Ando, S., Garzanti, E., Padoan, M., and Limonta, M., 2012, Corrosion of heavy minerals during 1061 weathering and diagenesis: a catalogue for optical analysis: Sedimentary Geology, v. 280, 1062 p. 165–178. 1063

Bassi, D., and Nebelsick, J. H., 2000, Calcareous algae from the lower Oligocene Gornji Grad 1064 beds of Northern Slovenia: Rivista Italiana di Paleontologia e Stratigrafia, v. 106, p. 1065 99-122. 1066

Bea, R. G., Sircar, P., and Niedoroda, A. W., 1983, Wave-induced slides in South Pass Block 70, 1067 Mississippi Delta: Journal of Geotechnical Engineering, v. 109, p. 619–644. 1068

Bhattacharya, G. C. B., Chaubey, A. K., Murty, G. P. S., Srinivas, S., Sarma, K. V., 1069 Subrahmanyam, V., and Krishna, K. S., 1994, Evidence for seafoor spreading in the 1070 Laxmi Basin, northeastern Indian Ocean: Earth and Planetary Science Letters, v. 125, p. 1071 211-220. 1072

Bilham, R., 1999, Slip parameters for the Rann of Kachchh, India, 16 June 1819, earthquake, 1073 quantified from contemporary accounts, in Stewart, I. S., and Vita-Finzi, C., eds., Coastal 1074 Tectonics, Volume 146: London, Geological Society, p. 295-319. 1075

Bilham, R., Bendick, R., and Wallace, K., 2003, Flexure of the Indian Plate and intraplate 1076 earthquakes: Proceedings of the Indian Academy of Sciences (Earth and Planetary 1077 Sciences), v. 112,, p. 1–14. 1078

Biswas, S. K., 2005, A review of structure and tectonics of Kutch basin, western India, with 1079 special reference to earthquakes: Current science, v. 88, no. 10, p. 1592-1600. 1080

BouDagher-Fadel, M., 2018, Evolution and geological significance of larger benthic 1081 foraminifera, London, UCL Press, 702 p. 1082

Bryn, P., Berg, K., Stoker, M. S., Haflidason, H., and Solheim, A., 2005, Contourites and their 1083 relevance for mass wasting along the Mid-Norwegian Margin.: Marine and Petroleum 1084 Geology, v. 22, p. 85-96. 1085

Burg, J. P., Bernoulli, D., Smit, J., Dolati, A., and Bahroudi, A., 2008, A giant catastrophic mud‐1086 and‐debris flow in the Miocene Makran: Terra Nova, v. 20, no. 3, p. 188-193. 1087

47

Callot, P., Sempere, T., Odonne, F., and Robert, E., 2008, Giant submarine collapse of a 1088 carbonate platform at the Turonian-Coniacian transition: the Ayabacas formation, 1089 southern Peru: Basin Research, v. 20, p. 333–357. 1090

Calvès, G., Huuse, M., Clift, P. D., and Brusset, S., 2015, Giant fossil mass wasting off the coast 1091 of West India: The Nataraja submarine slide: Earth and Planetary Science Letters, v. 432, 1092 p. 265–272. doi:10.1016/j.epsl.2015.10.022. 1093

Carracedo, J. C., 1999, Growth, structure, instability and collapse of Canarian volcanoes and 1094 comparisons with Hawaiian volcanoes: Journal of Volcanology and Geothermal 1095 Research, v. 94, no. 1-4, p. 1-19. 1096

Carter, A., and Bristow, C. S., 2003, Linking hinterland evolution and continental basin 1097 sedimentation by using detrital zircon thermochronology; a study of the Khorat Plateau 1098 basin, eastern Thailand: Basin Research, v. 15, p. 271–285. 1099

Carter, R. M., 1975, A discussion and classification of subaqueous mass-transport with particular 1100 application to grain-flow, slurry-flow, and fluxoturbidites: Earth-Science Reviews, v. 11, 1101 no. 2, p. 145-177. 1102

Chowdhary, L. R., 2004, Petroleum Geology of the Cambay Basin, Gujarat., India, Indian 1103 Petroleum Publishers, 171 p. 1104

Clift, P., Gaedicke, C., Edwards, R., Lee, J., II, Hildebrand, P., Amjad, S., White, R. S., and 1105 Schlueter, H.-U., 2002, The stratigraphic evolution of the Indus Fan and the history of 1106 sedimentation in the Arabian Sea: Marine Geophysical Researches, v. 23, no. 3, p. 223-1107 245. 1108

Clift, P. D., 2006, Controls on the erosion of Cenozoic Asia and the flux of clastic sediment to 1109 the ocean: Earth and Planetary Science Letters, v. 241, no. 3-4, p. 571-580. 1110

Clift, P. D., and Blusztajn, J. S., 2005, Reorganization of the western Himalayan river system 1111 after five million years ago: Nature, v. 438, no. 7070, p. 1001-1003. 1112

Clift, P. D., Campbell, I. H., Pringle, M. S., Carter, A., Zhang, X., Hodges, K. V., Khan, A. A., 1113 and Allen, C. M., 2004, Thermochronology of the modern Indus River bedload; new 1114 insight into the control on the marine stratigraphic record: Tectonics, v. 23, no. TC5013. 1115 doi:10.1029/2003TC001559. 1116

Clift, P. D., Giosan, L., Blusztajn, J., Campbell, I. H., Allen, C. M., Pringle, M., Tabrez, A., 1117 Danish, M., Rabbani, M. M., Carter, A., and Lückge, A., 2008a, Holocene erosion of the 1118 Lesser Himalaya triggered by intensified summer monsoon: Geology, v. 36, no. 1, p. 79–1119 82. doi: 10.1130/G24315A.1. 1120

Clift, P. D., Giosan, L., Carter, A., Garzanti, E., Galy, V., Tabrez, A. R., Pringle, M., Campbell, 1121 I. H., France-Lanord, C., Blusztajn, J., Allen, C., Alizai, A., Lückge, A., Danish, M., and 1122 Rabbani, M. M., 2010, Monsoon control over erosion patterns in the Western Himalaya: 1123 possible feed-backs into the tectonic evolution, in Clift, P. D., Tada, R., and Zheng, H., 1124 eds., Monsoon evolution and tectonic-climate linkage in Asia, Volume 342: London, 1125 Geological Society, p. 181–213. 1126

Clift, P. D., Hodges, K., Heslop, D., Hannigan, R., Hoang, L. V., and Calves, G., 2008b, Greater 1127 Himalayan exhumation triggered by Early Miocene monsoon intensification: Nature 1128 Geoscience, v. 1, p. 875-880. doi:10.1038/ngeo351. 1129

Clift, P. D., Shimizu, N., Layne, G., Gaedicke, C., Schlüter, H. U., Clark, M. K., and Amjad, S., 1130 2001, Development of the Indus Fan and its significance for the erosional history of the 1131 western Himalaya and Karakoram: Geological Society of America Bulletin, v. 113, p. 1132 1039–1051. 1133

48

Clift, P. D., Zhou, P., Stockli, D. F., and Blusztajn, J., 2018, An Erosional Record of Regional 1134 Exhumation of the Western Himalaya and Karakoram: Nature Communications, v. in 1135 review. 1136

Coleman, J., and Prior, D., 1988, Mass wasting on continental margins: Annual Review of Earth 1137 and Planetary Sciences, v. 16, no. 1, p. 101-119. 1138

Crawford, M. B., and Searle, M. P., 1992, Field relationships and geochemistry of pre-collisional 1139 (India-Asia) granitoid magmatism in the central Karakoram, northern Pakistan: 1140 Tectonophysics, v. 206, no. 1-2, p. 171-192. 1141

Dan, G., Sultan, N., and Savoye, B., 2007, The 1979 Nice harbour catastrophe revisited: trigger 1142 mechanism inferred from geotechnical measurements and numerical modelling: Marine 1143 Geology, v. 245, p. 40–64. 1144

Davies, R., and Clark, I., 2006, Submarine slope failure primed and triggered by silica and its 1145 diagenesis: Basin Research, v. 18, p. 339-350. 1146

DeCelles, P. G., Gehrels, G. E., Najman, Y., Martin, A. J., Carter, A., and Garzanti, E., 2004, 1147 Detrital geochronology and geochemistry of Cretaceous-Early Miocene strata of Nepal: 1148 implications for timing and diachroneity of initial Himalayan orogenesis: Earth and 1149 Planetary Science Letters, v. 227, no. 3-4, p. 313-330. 1150

DeCelles, P. G., Gehrels, G. E., Quade, J., LaReau, B., and Spurlin, M., 2000, Tectonic 1151 implications of U-Pb zircon ages of the Himalayan orogenic belt in Nepal: Science, v. 1152 288, no. 5465, p. 497-499. doi:10.1126/science.288.5465.497. 1153

DeCelles, P. G., Kapp, P., Gehrels, G. E., and Ding, L., 2014, Paleocene-Eocene foreland basin 1154 evolution in the Himalaya of southern Tibet and Nepal: Implications for the age of initial 1155 India-Asia collision: Tectonics, v. 33, p. 824–849. doi:10.1002/ 2014TC003522. 1156

Deniel, C., Vidal, P., Fernandez, A., Fort, P., and Peucat, J.-J., 1987, Isotopic study of the 1157 Manaslu granite (Himalaya, Nepal): inferences on the age and source of Himalayan 1158 leucogranites: Contributions to Mineralogy and Petrology, v. 96, no. 1, p. 78-92. 1159

Derry, L. A., and France-Lanord, C., 1996, Neogene Himalayan weathering history and river 1160 87Sr/86Sr; impact on the marine Sr record: Earth and Planetary Science Letters, v. 142, p. 1161 59-74. 1162

Donelick, R. A., O’Sullivan, P. B., and Ketcham, R. A., 2005, Apatite fission-track analysis: 1163 Reviews in Mineralogy and Geochemistry, v. 58, no. 1, p. 49-94. 1164

Dunham, R. J., 1962, Classification of carbonate rocks according to depositional texture, in Ham, 1165 W. E., ed., Classification of Carbonate Rocks, Volume 1, American Association of 1166 petroleum geology, p. 108–121. 1167

Dunlap, W. J., and Wysoczanski, R., 2002, Thermal evidence for Early Cretaceous 1168 metamorphism in the Shyok suture zone and age of the Khardung volcanic rocks, 1169 Ladakh, India: Journal of Asian Earth Sciences, v. 20, no. 5, p. 481-490. 1170

Ekdale, A. A., Bromley, R. G., and Pemberton, S. G., 1984, Ichnology. Trace fossils in 1171 sedimentation and stratigraphy, Tulsa, OK, Society of Economic Palaeontologists and 1172 Mineralogists, Short Course, 317 p. 1173

Elverhoi, A., Breien, H., Blasio, F. V. D., Harbitz, C. B., and Pagliardi, M., 2010, Submarine 1174 landslides and the importance of the initial sediment composition for run-out length and 1175 final deposit: Ocean Dynamics, v. 60, p. 1027–1046. doi 10.1007/s10236-010-0317-z. 1176

Embley, R. W., and Jacobi, R. D., 1977, Distribution and morphology of large submarine 1177 sediment slides and slumps on Atlantic continental margins: Marine Georesources & 1178 Geotechnology, v. 2, no. 1-4, p. 205-228. 1179

49

Evans, D., McGiveron, S., Harrison, Z., Bryn, P., and Berg, K., 2002, Along-slope variation in 1180 the late Neogene evolution of the mid-Norwegian margin in response to uplift and 1181 tectonism, in Dore, A. G., Cartwright, J. A., Stoker, S., Turner, J. P., and White, N., eds., 1182 Exhumation of the North Atlantic Margin: Timing, Mechanisms and Implications for 1183 Petroleum Exploration, Volume 196: London, Geological Society, p. 139-151. 1184

Farrell, S. G., 1984, A dislocation model applied to slump structures, Ainsa Basin, South Central 1185 Pyrenees: Journal of Structural Geology, v. 6, p. 727-736. 1186

Fedo, C. M., Nesbitt, H. W., and Young, G. M., 1995, Unraveling the effects of potassium 1187 metasomatism in sedimentary rocks and paleosols, with implications for paleoweathering 1188 conditions and provenance: Geology, v. 23, p. 921–924. 1189

Fraser, J. E., Searle, M. P., Parrish, R. R., and Noble, S. R., 2001, Chronology of deformation, 1190 metamorphism, and magmatism in the southern Karakoram Mountains: Geological 1191 Society of America Bulletin, v. 113, no. 11, p. 1443-1455. 1192

Frey Martinez, J., Cartwright, J., and Hall, B., 2005, 3D seismic interpretation of slump 1193 complexes: examples from the continental margin of Israel: Basin Research, v. 17, p. 83-1194 108. 1195

Gamboa, D., Alves, T., and Cartwright, J., 2012, Seismic-Scale Rafted and Remnant Blocks over 1196 Salt Ridges in the Espírito Santo Basin, Brazil, in Yamada, Y., Kawamura, K., Ken 1197 Ikehara, Ogawa, Y., Urgeles, R., Mosher, D., Chaytor, J., and Strasser, M., eds., 1198 Submarine Mass Movements and Their Consequences: Dordrecht, Springer, p. 629-638. 1199

Garzanti, E., 2017, The maturity myth in sedimentology and provenance analysis: Journal of 1200 Sedimentary Research, v. 87, p. 353-365. 1201

Garzanti, E., and Andò, S., 2007, Heavy-mineral concentration in modern sands: implications for 1202 provenance interpretation, in Mange, M., and Wright, D., eds., Heavy Minerals in Use, 1203 Volume 58: Amsterdam, Elsevier, p. 517-545. 1204

Garzanti, E., Andò, S., and Vezzoli, G., 2009, Grain-size dependence of sediment composition 1205 and environmental bias in provenance studies: Earth and Planetary Science Letters, v. 1206 277, no. 3-4, p. 422-432. 1207

Garzanti, E., Vezzoli, G., Ando, S., Paparella, P., and Clift, P. D., 2005, Petrology of Indus River 1208 sands; a key to interpret erosion history of the western Himalayan syntaxis: Earth and 1209 Planetary Science Letters, v. 229, no. 3-4, p. 287-302. DOI: 10.1016/j.epsl.2004.11.008. 1210

Gehrels, G. E., Kapp, P., DeCelles, P., Pullen, A., Blakely, R., Weislgel, A., Ding, L., Guynn, J., 1211 Marin, A., McQuarrie, N., and Yin, A., 2011, Detrital zircon geochronology of pre-1212 Tertiary strata in the Tibetan-Himalayan orogen: Tectonics, v. 30, no. TC5016. 1213 doi:10.1029/2011TC002868. 1214

Goldstein, S. L., O'Nions, R. K., and Hamilton, P. J., 1984, A Sm-Nd isotopic study of 1215 atmospheric dusts and particulates from major river systems: Earth and Planetary Science 1216 Letters, v. 70, no. 2, p. 221-236. 1217

Goswami, V., Singh, S. K., Bhushan, R., and Rai, V. K., 2012, Temporal variations in 87Sr/86Sr 1218 and eNd in sediments of the southeastern Arabian Sea: Impact of monsoon and surface 1219 water circulation: Geochemistry Geophysics Geosystems, v. 13, no. Q01001. 1220 doi:10.1029/2011GC003802. 1221

Gradstein, F. M., Ogg, G., and Schmitz, M., 2012, The Geologic Time Scale 2012, New York, 1222 Elsevier. 1223

Hampton, M. A., Lee, H. J., and Locat, J., 1996, Submarine landslides: Reviews of Geophysics, 1224 v. 34, no. 1, p. 33-59. doI:10.1029/95RG03287 1225

50

Herron, M. M., 1988, Geochemical classification of terrigenous sands and shales from core or 1226 log data: Journal of Sedimentary Petrology, v. 58, p. 820–829. 1227

Hine, A. C., 1992, Megabreccia shedding from modern, low-relief carbonate platforms, 1228 Nicaraguan Rise: Geological Society of America Bulletin, v. 104, p. 928–943. 1229

Hodges, K., 2003, Geochronology and thermochronology in orogenic systems, in Rudnick, R., 1230 ed., The Crust: Amsterdam, Elsevier-Science, p. 263-292. 1231

Honegger, K., Dietrich, V., Frank, W., Gansser, A., Thoni, M., and Trommsdorf, V. F., 1982, 1232 Magmatism and metamorphism in the Ladakh Himalayas (The Indus-Tsangpo suture 1233 zone): Earth and Planetary Science Letters, v. 60, p. 178-194. 1234

Hsu, S.-K., Kuo, J., Lo, C.-L., Tsai, C.-H., Doo, W.-B., Ku, C.-Y., and Sibuet, J.-C., 2008, 1235 Turbidity currents, submarine landslides and the 2006 Pingtung earthquake off SW 1236 Taiwan: Terrestrial Atmospheric and Ocean Sciences, v. 19, no. 6, p. 767–772. . 1237 doi:10.3319/TAO.2008.19.6.767(PT). 1238

Hubert, J. F., 1962, A zircon–tourmaline–rutile maturity index and the interdependence of 1239 the composition of heavy mineral assemblages with the gross composition and texture of 1240 sandstones: Journal of Sedimentary Petrology, v. 32, p. 440-450. 1241 Inger, S., and Harris, N., 1993, Geochemical constraints on leucogranite magmatism in the 1242

Langtang Valley, Nepal Himalaya: Journal of Petrology, v. 34, no. 2, p. 345-368. 1243 Kastens, K. A., 1984, Earthquakes as a triggering mechanism for debris flows and turbidites on 1244

the Calabrian Ridge: Marine Geology, v. 55, no. 1-2, p. 13-33. 1245 Khan, M. A., Stern, R. J., Gribble, R. F., and Windley, B. F., 1997, Geochemical and isotopic 1246

constraints on subduction polarity, magma sources, and palaeogeography of the Kohistan 1247 intra-oceanic arc, northern Pakistan Himalaya: Journal of the Geological Society, 1248 London, v. 154, p. 935–946. 1249

Khonde, N., Singh, S. K., Maurya, D. M., Rai, V. K., Chamyal, L. S., and Giosan, L., 2017, 1250 Tracing the Vedic Saraswati River in the Great Rann of Kachchh: Scientific Reports, v. 1251 7, no. 5476. doi:10.1038/s41598-017-05745-8. 1252

Klaus, A., Norris, R. D., Kroon, D., and Smit, J., 2000, Impact-induced mass wasting at the KT 1253 boundary: Blake Nose, western North Atlantic: Geology, v. 28, no. 4, p. 319-322. 1254

Krastel, S., Wynn, R. B., Georgiopoulou, A., Geersen, J., Henrich, R. d., Meyer, M., and 1255 Schwenk, T., 2012, Large-Scale Mass Wasting on the Northwest African Continental 1256 Margin: Some General Implications for Mass Wasting on Passive Continental Margins, 1257 in Yamada, Y., Kawamura, K., Ken Ikehara, Ogawa, Y., Urgeles, R., Mosher, D., 1258 Chaytor, J., and Strasser, M., eds., Submarine Mass Movements and Their Consequences: 1259 Dordrecht, Springer, p. 189-200. 1260

Krol, M. A., Zeitler, P. K., and Copeland, P., 1996, Episodic unroofing of the Kohistan 1261 Batholith, Pakistan: Implications from K-feldspar thermochronology: Journal of 1262 Geophysical Research-Solid Earth, v. 101, no. B12, p. 28149-28164. 1263

Kurian, S., Nath, B. N., Kumar, N. C., and Nair, K. K. C., 2013, Geochemical and Isotopic 1264 Signatures of Surficial Sediments from the Western Continental Shelf of India: Inferring 1265 Provenance, Weathering, and the Nature of Organic Matter Geochemical and Isotopic 1266 Signatures of Sediments From The Indian West Coast: Journal of Sedimentary Research, 1267 v. 83, no. 6, p. 427-442. doi:10.2110/jsr.2013.36. 1268

Kusznir, N. J., Roberts, A. M., and Morley, C. K., 1995, Forward and reverse modelling of rift 1269 basin formation, in Lambiase, J. J., ed., Hydrocarbon Habitat in Rift Basins, Volume 80: 1270 London, Geological Society, p. 33–56. 1271

51

Lastras, G., Canals, M., Urgeles, R., De Batist, M., Calafat, A. M., and Casamor, J. L., 2004, 1272 Characterisation of the recent BIG’95 debris flow on the Ebro margin, Western 1273 Mediterranean Sea, after a variety of seismic reflection data: Marine Geology, v. 213, p. 1274 235-255. doi:10.1016/j.margeo.2004.10.008. 1275

Le Fort, P., Debon, F., and Sonet, J., 1983, Petrography, geochemistry and geochronology of 1276 some samples from the Karakoram Batholith (N. Pakistan), in Shams, F. A., ed., Granites 1277 of the Himalayas, Karakoram and Hindu Kush: Lahore, Pakistan, Punjab University, p. 1278 377–387. 1279

Leynaud, D., Mienert, J., and Vanneste, M., 2009, Submarine mass movements on glaciated and 1280 non-glaciated European continental margins: A review of triggering mechanisms and 1281 preconditions to failure: Marine and Petroleum Geology, v. 26, no. 5, p. 618-632. 1282

Li, Y., 2018, Signal propagation, provenance, and climate records in the Indus Submarine 1283 Canyon since the Last Deglacial [PhD PhD]: Louisiana State University. 1284

Li, Y., Clift, P. D., Böning, P., Blusztajn, J., Murray, R. W., Ireland, T., Pahnke, K., and Giosan, 1285 L., 2018, Continuous Signal Propagation in the Indus Submarine Canyon since the Last 1286 Deglacial: Marine Geology, v. 406, p. 159–176. doi:10.1016/j.margeo.2018.09.011. 1287

Lowe, D. R., 1976, Grain flow and grain flow deposits: Journal of Sedimentary Research, v. 46, 1288 no. 1. 1289

Lucente, C. C., and Pini, G. A., 2003, Anatomy and emplacement mechanism of a large 1290 submarine slide within a Miocene foredeep in the northern Apennines, Italy: a field 1291 perspective: American Journal of Science, v. 303, p. 565-602. 1292

Mange, M. A., and Maurer, H. F. W., 1992, Heavy minerals in colour, London, Chapman and 1293 Hall, 147 p. 1294

Martinsen, O. J., and Bakken, B., 1990, Extensional and compressional zones in slumps and 1295 slides in the Namurian of County Clare, Ireland: Journal of the Geological Society of 1296 London, v. 147, p. 153-164. 1297

Mienert, J., Vanneste, M., Bünz, S., Andreassen, K., Haflidason, H., and Sejrup, H. P., 2005, 1298 Ocean warming and gas hydrate stability on the mid-Norwegian margin at the Storegga 1299 Slide: Marine and Petroleum Geology, v. 22, no. 1-2, p. 233-244. 1300

Miles, P. R., and Roest, W. R., 1993, Earliest seafoor spreading magnetic anomalies in the north 1301 Arabian Sea and the ocean-continent transition: Geophysical Journal International, v. 1302 115, p. 1025-1031. 1303

Moernaut, J., De Batist, M., Charlet, F., Heirman, K., Chapron, E., Pino, M., Brümmer, R., and 1304 Urrutia, R., 2007, Giant earthquakes in South-Central Chile revealed by Holocene mass-1305 wasting events in Lake Puyehue: Sedimentary Geology, v. 195, no. 3-4, p. 239-256. 1306

Moore, D. G., 1961, Submarine slumps: Journal of Sedimentary Research, v. 31, no. 3, p. 343-1307 357. 1308

Mullins, H. T., Gardulski, A. F., and Hine, A. C., 1986, Catastrophic collapse of the west Florida 1309 carbonate platform margin: Geology v. 14, p. 167–170. 1310

Nair, N., and Pandey, D. K., 2018, Cenozoic sedimentation in the Mumbai Offshore Basin: 1311 Implications for tectonic evolution of the western continental margin of India: Journal of 1312 Asian Earth Sciences, v. 152, p. 132-144. doi:10.1016/j.jseaes.2017.11.037. 1313

Najman, Y., Appel, E., Boudagher‐Fadel, M., Bown, P., Carter, A., Garzanti, E., Godin, L., Han, 1314 J., Liebke, U., Oliver, G., Parrish, R., and Vezzoli, G., 2010, Timing of India‐Asia 1315 collision: Geological, biostratigraphic, and palaeomagnetic constraints: Journal of 1316 Geophysical Research, v. 115, no. B12416. doi:10.1029/2010JB007673. 1317

52

Nesbitt, H. W., Markovics, G., and Price, R. C., 1980, Chemical processes affecting alkalis and 1318 alkaline earths during continental weathering: Geochimica et Cosmochimica Acta, v. 44, 1319 p. 1659–1666. 1320

Pandey, D. K., Clift, P. D., Kulhanek, D. K., Andò, S., Bendle, J. A. P., Bratenkov, S., Griffith, 1321 E. M., Gurumurthy, G. P., Hahn, A., Iwai, M., Khim, B.-K., Kumar, A., Kumar, A. G., 1322 Liddy, H. M., Lu, H., Lyle, M. W., Mishra, R., Radhakrishna, T., Routledge, C. M., 1323 Saraswat, R., Saxena, R., Scardia, G., Sharma, G. K., Singh, A. D., Steinke, S., Suzuki, 1324 K., Tauxe, L., Tiwari, M., Xu, Z., and Yu, Z., 2016a, Site U1456, in Pandey, D. K., Clift, 1325 P. D., and Kulhanek, D. K., eds., Arabian Sea Monsoon. Proceedings of the International 1326 Ocean Discovery Program, Volume 355: College Station, TX, International Ocean 1327 Discovery Program. 1328

-, 2016b, Site U1457, in Pandey, D. K., Clift, P. D., and Kulhanek, D. K., eds., Arabian Sea 1329 Monsoon. Proceedings of the International Ocean Discovery Program, Volume 355: 1330 College Station, TX, International Ocean Discovery Program. 1331

Pandey, D. K., Clift, P. D., Kulhanek, D. K., and Expedition 355 Scientists, 2016c, Arabian Sea 1332 Monsoon: Proceedings of the International Ocean Discovery Program, v. 355. 1333 doi:10.14379/iodp.proc.355.2016. 1334

Pandey, O. P., Agrawal, P. K., and Negi, J. G., 1995, Lithospheric structure beneath Laxmi 1335 Ridge and late Cretaceous geodynamic events: Geo-Marine Letters, v. 15, p. 85-91. 1336

Parnell, J., 2008, Global mass wasting at continental margins during Ordovician high meteorite 1337 influx: Nature Geoscience, v. 2, p. 57. doi:10.1038/ngeo386. 1338

Parrish, R. R., and Hodges, K. V., 1996, Isotopic constraints on the age and provenance of the 1339 Lesser and Greater Himalayan sequences, Nepalese Himalaya: Geological Society of 1340 America Bulletin, v. 108, no. 7, p. 904-911. 1341

Parrish, R. R., and Tirrul, R., 1989, U-Pb age of the Baltoro Granite, Northwest Himalaya, and 1342 implications for monazite U-Pb systematics: Geology, v. 17, p. 1076–1079. 1343

Peters, S. E., and Loss, D. P., 2012, Storm and fair-weather wave base: A relevant distinction?: 1344 Geology, v. 40, no. 6, p. 511-514. 1345

Pickering, K., Stow, D., Watson, M., and Hiscott, R., 1986, Deep-water facies, processes and 1346 models: a review and classification scheme for modern and ancient sediments: Earth-1347 Science Reviews, v. 23, no. 2, p. 75-174. 1348

Piper, D. J., Shor, A. N., Farre, J. A., O'Connell, S., and Jacobi, R., 1985, Sediment slides and 1349 turbidity currents on the Laurentian Fan: Sidescan sonar investigations near the epicenter 1350 of the 1929 Grand Banks earthquake: Geology, v. 13, no. 8, p. 538-541. 1351

Principaud, M., Mulder, T., Gillet, H., and Borgomano, J., 2015, Large-scale carbonate 1352 submarine mass-wasting along the northwestern slope of the Great Bahama Bank 1353 (Bahamas): Morphology, architecture, and mechanisms: Sedimentary Geology, v. 317, p. 1354 27-42. doi:10.1016/j.sedgeo.2014.10.008. 1355

Rao, R. P., and Talukdar, S. N., 1980, Petroleum geology of Bombay High field, India, in 1356 Halbouty, M. T., ed., Giant oil and gas fields of the decade: 1968–1978, Volume 30, 1357 American Association of Petroleum Geologists, p. 487–506. 1358

Rao, V. P., and Rao, B. R., 1995, Provenance and distribution of clay minerals in the sediments 1359 of the western continental shelf and slope of India: Continental Shelf Research, v. 15, no. 1360 14, p. 1757-1771. doi:10.1016/0278-4343(94)00092-2. 1361

53

Ravikant, V., Wu, F. Y., and Ji, W. Q., 2009, Zircon U-Pb and Hf isotopic constraints on 1362 petrogenesis of the Cretaceous-Tertiary granites in eastern Karakoram and Ladakh, India: 1363 Lithos, v. 110, p. 153-166. 1364

Rolland, Y., Picard, C., Pecher, A., Lapierre, H., Bosch, D., and Keller, F., 2002, The Cretaceous 1365 Ladakh arc of NW himalaya—slab melting and melt–mantle interaction during fast 1366 northward drift of Indian Plate: Chemical Geology, v. 182, p. 139–178. 1367

Schärer, U., Copeland, P., Harrison, T. M., and Searle, M. P., 1990, Age, cooling history, and 1368 origin of post-collisional leucogranites in the Karakoram Batholith; a multi-system 1369 isotope study: The Journal of Geology, v. 98, no. 2, p. 233-251. 1370

Schärer, U., Xu, R.-H., and Allègre, C. J., 1984, U-Pb geochronology of Gangdese 1371 (Transhimalaya) plutonism in the Lhasa-Xigaze region, Tibet: Earth and Planetary 1372 Science Letters, v. 69, no. 2, p. 311-320. 1373

Schlager, W., Reijmer, J. J. G., and Droxler, A., 1994, Highstand shedding of carbonate 1374 platforms: Journal of Sedimentary Research, v. 64, no. 3b, p. 270-281. 1375 doi:10.1306/D4267FAA-2B26-11D7-8648000102C1865D. 1376

Sclater, J. G., and Christie, P. A. F., 1980, Continental stretching: an explanation of the post 1377 Mid-Cretaceous subsidence of the central North Sea basin: Journal of Geophysical 1378 Research, v. 85, p. 3711–3739. 1379

Seilacher, A., 1967, Bathymetry of trace fossils: Marine Geology, v. 5, p. 413–428. 1380 Singh, M., Sharma, M., and Tobschall, H. J., 2005, Weathering of the Ganga alluvial plain, 1381

northern India: implications from fluvial geochemistry of the Gomati River: Applied 1382 Geochemistry, v. 20, p. 1-21. 1383

Singh, S., Kumar, R., Barley, M. E., and Jain, A. K., 2007, SHRIMP U–Pb ages and depth of 1384 emplacement of Ladakh batholith, eastern Ladakh, India: Journal of Asian Earth 1385 Sciences, v. 30, no. 3, p. 490–503. 1386

Singh, S. K., and France-Lanord, C., 2002, Tracing the distribution of erosion in the 1387 Brahmaputra watershed from isotopic compositions of stream sediments: Earth and 1388 Planetary Science Letters, v. 202, no. 3-4, p. 645-662. 1389

Strachan, L. J., 2002, Slump-initiated and controlled syndepositional sandstone remobilisation: 1390 an example from the Namurian of County Clare, Ireland: Sedimentology, v. 49, p. 25-41. 1391

Talling, P. J., Masson, D. G., Sumner, E. J., and Malgesini, G., 2012, Subaqueous sediment 1392 density flows: Depositional processes and deposit types: Sedimentology, v. 59, no. 7, p. 1393 1937-2003. 1394

Tappin, D. R., Watts, P., McMurtry, G. M., Lafoy, Y., and Matsumoto, T., 2001, The Sissano, 1395 Papua New Guinea tsunami of July 1998 – offshore evidence on the source mechanism: 1396 Marine Geology, v. 17, p. 1–23. 1397

Trincardi, F., and Argnani, A., 1990, Gela Submarine slide: a major basin-wide event in the Plio-1398 Quatemary foredeep of Sicily: Geo-marine Letters, v. 10, p. 13-21. 1399

Vardy, M. E., Pinson, L. J., Bull, J. M., Dix, J. K., Henstock, T. J., Davis, J. W., and Gutowski, 1400 M., 2010, 3D seismic imaging of buried Younger Dryas mass movement flows: Lake 1401 Windermere, UK: Geomorphology, v. 118, no. 1-2, p. 176-187. 1402

Vermeesch, P., 2004, How many grains are needed for a provenance study?: Earth and Planetary 1403 Science Letters, v. 224, p. 351–441. 1404

Wandrey, C. J., 2004, Bombay Geologic Province Eocene to Miocene Composite Total 1405 Petroleum System, India: US Geological Survey, Department of the Interior, 2208-F. 1406

54

Weinberg, R. F., and Dunlap, W. J., 2000, Growth and Deformation of the Ladakh Batholith, 1407 Northwest Himalayas: Implications for Timing of Continental Collision and Origin of 1408 Calc‐Alkaline Batholiths: Journal of Geology, v. 108, p. 303–320. DOI: 10.1086/314405. 1409

Yamada, Y., Kawamura, K., Ikehara, K., Ogawa, Y., Urgeles, R., Mosher, D., Chaytor, J., and 1410 Strasser, M., 2012, Submarine mass movements and their consequences, in Lykousis, V., 1411 Sakellariou, D., and Locat, J., eds., Submarine mass movements and their consequences, 1412 Springer, p. 1-12. 1413

Zeilinger, G., Burg, J. P., Schaltegger, U., and Seward, D., 2001, New U/ Pb and fission track 1414 ages and their implication for the tectonic history of the lower Kohistan Arc Complex, 1415 northern Pakistan: Journal of Asian Earth Sciences, v. 19, no. 3S, p. 79-81. 1416

Zeitler, P. K., and Chamberlain, C. P., 1991, Petrogenetic and Tectonic Significance of Young 1417 Leukogranites from the Northwestern Himalaya, Pakistan: Tectonics, v. 10, no. 4, p. 729-1418 741. 1419

Zeitler, P. K., Chamberlain, C. P., and Smith, H. A., 1993, Synchronous Anatexis, 1420 Metamorphism, and Rapid Denudation at Nanga-Parbat (Pakistan Himalaya): Geology, v. 1421 21, no. 4, p. 347-350. 1422

Zhuang, G., Najman, Y., Guillot, S., Roddaz, M., Antoine, P.-O., Métaise, G., Carter, A., 1423 Marivaux, L., and Solangig, S. H., 2015, Constraints on the collision and the pre-collision 1424 tectonic configuration between India and Asia from detrital geochronology, 1425 thermochronology, and geochemistry studies in the lower Indus basin, Pakistan: Earth 1426 and Planetary Science Letters, v. 432, p. 363–373. doi:10.1016/j.epsl.2015.10.026. 1427

1428

Figure 1Dailey et al.

Bay of B

engalArabian Sea

Laxmi Ridge

222

223

721722731

U1456

720

219

Laccadive Ridge

Deccan Plateau

Carlsberg Ridge

ARABIA

Karakoram

Hindu K

ush

Indusdelta

36ºN

34ºN

32ºN

30ºN

28ºN

26ºN

24ºN

22ºN

20ºN

18ºN

16ºN

14ºN

12ºN

10ºN

54ºE 58ºE 62ºE 66ºE 70ºE 74ºE 78ºE 82ºE 86ºE

U1457

23

24

25

26

27

26

24

23

?

?

?

?

?

?

??

?

Indo-Gangetic Plain

LaxmiBasin

Gulf of Oman

MakranIn

dus

Rive

r

Western G

hats

Muscat

Karachi

Goa

Murray Ridg

e

N armada

50004000

01000

-1000-1500-2000-2500-3000

2000

Topography (m)

Tapti

221

Mahi

Fig. 18

SiteU1456

Laxmi Ridge

Laxmi

Basin

3500

30002500

1500500

Indian Shelf

27

26

WadiaGuyot

RamanGuyot

Cam

bay

Basi

nGujarat-

Saurashtra Margin

Rann ofKutch

Bombay High

Bombay High Nanga Parbat

Sula

iman

Kirth

ar

Lesser Himalaya

Greater Himalaya

Transhimalaya

Mumbai

AB

Site U1457

Fig. 2A

Fig. 2B

Gulf ofCambay

Figure 1

Shotpoint2000 2100 2200 2300 2400 2500 2600

5.0

5.5

6.0

TWT

(s)

Shotpoint2000 2100 2200 2300 2400 2500 2600

(A) Site U1456

(B) Site U1457

11424 11585 11746 11907 12068

U1456

11424 11585 11746 11907 12068

6.0

5.5

TWT

(s)

Shotpoint11424 11585 11746 11907 12068

Shotpoint

Basement

Paleocene-Middle Miocene

MTC Phase 2 (Nataraja Slide)

Upper Miocene

MTC Phase 1 (Nataraja Slide)Pliocene

Upper Miocene-Pliocene

Lower Pleistocene

MTC

MTC

Figure 2Dailey et al.

Figure 2

Dep

th (m

bsf)

Figure 3Dailey et al.

ClaySilt

SandCongl.

800

850

900

950

1000

1050

1100

(A) Site U1456

Debris Flow

Coherent Slide

Slump

Turbidites

Hemipelagic

1100

1050

1000

950

900

850

ClaySilt

SandCong

Sedimentary top

MTC Base

(B) Site U1457

Volcaniclastic

Dominant lithology and facies Limestone

Clay/Claystone

Silt/Siltstone

Sand/Sandstone

Gravel

Calcarenite

Carbonate Breccia

Volcanic Rock

Biostratigraphictop

Phase 1

Phase 2

Biostratigraphic top

Figure 6B

Figure 4A

Sedimentary top

Figure 4B

Figure 5A

Figu

re 5

B

MTC Base

Recov

ery

Recov

ery

Facies

Facies

Lithology

Lithology

Fig. 7A

Fig. 7B Fig. 8AFig. 8B+C

Fig. 8D

Fig. 9A

Fig. 9B

Fig. 9C

Fig. 9D

Fig. 6A HemipelagicSedimentation

Fig.10A+12B,E

Fig. 10B

Fig. 10C

Fig. 10D

Fig.11A

Fig.11BFig.11D

Fig.12A

Fig.12FFig.12D

Zircon U-Pb sample

Fig.12C

Fig.12GFig. 12H,I

Figure 3

800

840

34R

35R

36R

37R

38R

39R

40R

41R

Dep

th (m

bsf)

33R760

820

780

Lithology and sedimentary structuresCarbonate ooze/stone

Clay/Claystone

Silt/Siltstone

Calcarenite Parallel Laminations

Burrows

Wavy Laminations

Slump Fold

Small Scale Fault

Tilted Bedding

Breccia

Sand/Sandstone

Sedimentarytop of the MTC

Figure 4Dailey et al.

42R

Conglom.

ClaySilt Sand

CMF860

860

90070

R71

R69

R

880

ClaySilt Sand

CMF940

72R

73R

840

Sedimentarytop of the MTC

920

74R

75R

76R

77R

78R

(A) Site U1456 (B) Site U1457

Biostratigraphictop of the MTC

Zirconsample

Biostratigraphic top of the MTC at 1009 mbsf

Seismictop

Recov

ery

Recov

ery

Fig. 10C

Fig.11A

Fig.11B

Fig.11D

Figure 4

Dep

th (m

bsf)

Figure 5Dailey et al.

1105

17R

18R

16R

1085

107519

R

1095

Conglom.

ClaySilt Sand

CMF

Sedimentarybottom of the MTC

91R

92R

1050

1040

86R

1030

87R

88R1020

89R

1010

90R

Conglom.

ClaySilt Sand

CMF

Sedimentarybottom of the MTC

(A) Site U1456 (B) Site U1457

Recov

ery

Recov

ery

Lithology Lithology

Fig. 8D

Fig. 8A

Fig. 9C

Fig. 10D

Fig.12C,F

Figure 5

50R

Dep

th (m

bsf)

918

Figure 6Dailey et al.

51R Top of

hemipelagiclayer

938

ClaySilt Sand

CMF

922

930

926

934

942

946

52R

950

954

53R

(A) Site U1456 (B) Site U1456

980

990

57R

58R

985

975

59R

1000

Conglom.

ClaySilt Sand

CMF

56R

995

6R

Top of 1st carbonate pulse

Recov

ery

Lithology

Recov

ery

Lithology

Figs. 10A+12B

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Figure 11

Figure 12

log

(Fe 2

O3/K

2O)

1

0

-1210.5 1.5

Fe shaleFe sand

Shale

Wac

ke

Lith-arenite

Arkose

Sublith-arenite

Subarkose

Quartzarenite

B)

log (SiO2/Al2O3)

Indusdelta

Induscanyon

Figure 13Dailey et al.

ksp

bt

il

kao, chl, gb

smm

pl

K2OCaO+Na2O

CIA

100

90

80

70

60

50

40

Al2O3

Indus delta

Induscanyon

A)

ksp

Western Indian ShelfPaleoceneHemipelagic UnitPhase 2Phase 1 (U1456)Phase 1 (U1457)Phase 1 upper (U1457)

Figure 13

Nar

mad

a

Tapt

i

Tapt

i

Nar

mad

a

Indu

s

Nar

mad

aTapt

i

A) Site U1456 87Sr/86SrEpsilon Nd

Sediment-basedslide top

Seismic-basedslide top

B) Site U1457 Epsilon Nd

Slide base

Figure 14Dailey et al.

Dep

th (m

bsf)

ClaySilt

SandCongl.

800

850

900

950

1000

1050

1100

850

900

950

1000

1050

1100Clay

SiltSand

Congl.

-10 -6 -2-4-12 -8 0 0.71 0.730.72

Dep

th (m

bsf)

-10 -6 -2-4-12 -8 0

Slide base

Sediment-basedslide top

Deccan?

Tapt

i

Nar

mad

a

0.71 0.730.72 20 40 60 80 100

20 40 60 80 100

Clay (%)

87Sr/86Sr Clay (%)

SmectiteIllite+Chlorite

SmectiteIllite+Chlorite

Phase 2

Phase 1

Indu

sIn

dus

Figure 14

0.70 0.71 0.72 0.73 0.74 0.75

0

-10

-20

-5

-15

87Sr/86Sr

Transhimalaya

GreaterHimalaya

Nd

<15 ka Indus delta

Sites U1456 & U1457 fan

Mahi

Narmada

Nataraja MTC

Figure 15Dailey et al.

Peninsular rivers

Deccan

KirtharSulaimanforeland

Tapti

Quat. Indus canyon

Karakoram

Figure 15

ModernIndus

Indus Fan 7.8 Ma

Indus Fan 8.3 Ma

Indus Fan 15.6 Ma

U1457C-71RNataraja MTC

TethyanHimalaya

GreaterHimalaya

Karakoram

LesserHimalaya

Kohistan

n=96

n=133

n=121

n=119

n=51

n=3912

n=822

n=259

n=200

0 1000 2000 3000 4000

Age [Ma]

n=1084

Figure 16Dailey et al.

Karako

ram/

Kohist

anTeth

yan-G

reater

Himala

yaLe

sser

Himala

ya

0 20 40 60 80 100

Deccan Plateau

Age [Ma]

U1456C-71RNataraja MTC

Figure 16

Illite+Chlorite

SmectiteKaolinite

Figure 17Dailey et al.

Indus CanyonIndus shelf/deltaIndus floodplains

W. Indian Shelf

SaurashtraShelf

Western Indian ShelfPaleoceneHemipelagic UnitPhase 2Phase 1 (U1456)Phase 1 (U1457)Phase 1 upper (U1457)

Indus Fan

Gulf ofCambay

Figure 17

Age (Ma)0 10 20 30 40 50 60 70

Sedi

men

tatio

n R

ate

(km

3 /km

of m

argi

n/m

.y.)

0

20

40

60

80

Figure 18Dailey et al

0

1

2

3

4

5

6

70 50 100 150 200 250

Distance (km)

Dep

th (k

m)

SeafloorTop MioceneTop M. MioceneTop L. Miocene

Top L. OligoceneTop L. EoceneTop U. PaleoceneBasement

A)

B)

Fan-SlopeBoundary

Figure 18

penedoversteepgmargin

ak layersecond weaein carbonate

pplatformincippi ient slide

turbidite

incipppi ient slide

plankton yerweak laynatein carbon

pplatform

1Erosion of the underlying sediment as carbonate block moves downslope and begins to break up

First phase continues downslope and begins to turn into debris flow at front of the carbonate blocks

2

Hemipelagic rain deposited onto first phase

3

5 Smaller, secondphase movesdownslope

Continued slumping followingfirst phase

4

Figure 19Dailey et al

A B

C D

Figure 19

Sticky Note

It's difficult to accept that you have precision to two significant figures. in the younger grains I could believe one sig. fig., but the industry standard is a minimum of 1% analysis error, and thus the 100 Ma grains cannot be as precise as x.xx. Indeed they should be listed as xxx Ma +/- x Ma (or xxxx Ma +/- xx Ma).